Skip to content

Instantly share code, notes, and snippets.

@fonnesbeck

fonnesbeck/ARI Severity Analysis.ipynb

Created February 11, 2014 02:17
Show Gist options
  • Save fonnesbeck/8928121 to your computer and use it in GitHub Desktop.
Save fonnesbeck/8928121 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
ARI Severity Analysis.ipynb
{
"metadata": {
"name": "",
"signature": "sha256:fd86146639968f3c8b361f0274fdcbae8c8682d0d0166b8416345088bae3e280"
},
"nbformat": 3,
"nbformat_minor": 0,
"worksheets": [
{
"cells": [
{
"cell_type": "code",
"collapsed": false,
"input": [
"%matplotlib inline\n",
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import pandas as pd\n",
"import plotly"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 6
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Correlate RSV severity with vitamin D levels (at hospital)\n",
"\n",
"- severity = oxygen, mech. vent, ICU, length of stay, death\n",
"\n",
"Covariates:\n",
"\n",
"- Prematurity\n",
"- low vitamin D\n",
"- Congenital heart disease\n",
"- any previous medical condition \n",
"- Smoke exposure\n",
"- And maternal smoking during pregnancy.\n",
"- <10 weeks of age at time of season (or we can just say younger age if this is better)\n",
"- Daycare\n",
"- Siblings school age\n",
"- Low birth weight\n",
"- Male sex\n",
"- high viral load (low Ct number)\n",
"- young maternal age\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Set flag for extracting the RSV-positive subset"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"RSV_only = False"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 7
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized = pd.read_csv('data/hospitalized.csv', index_col=0)\n",
"hospitalized.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>greater_48hrs</th>\n",
" <th>fever_neutropenia</th>\n",
" <th>never_left</th>\n",
" <th>written_consent</th>\n",
" <th>child_name</th>\n",
" <th>mother_name</th>\n",
" <th>mother_birth_date</th>\n",
" <th>mother_record</th>\n",
" <th>mother_nationality</th>\n",
" <th>other_mother_nationality</th>\n",
" <th>father_nationality</th>\n",
" <th>other_father_nationality</th>\n",
" <th>child_record</th>\n",
" <th>newborn_id</th>\n",
" <th>newborn_id_number</th>\n",
" <th>adm_ari</th>\n",
" <th>adm_apnea</th>\n",
" <th>adm_asthma</th>\n",
" <th>adm_bronchiolitis</th>\n",
" <th>adm_bronchopneumo</th>\n",
" <th></th>\n",
" </tr>\n",
" <tr>\n",
" <th>case_id</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A0001</th>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> Remas Mahmoud Jbarah</td>\n",
" <td> Huda Katalo</td>\n",
" <td> 1976-01-21</td>\n",
" <td> NaN</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 2002048926</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0002</th>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> Majed Abdel Kareem Majed</td>\n",
" <td> Noor SHa'aban Mahmood</td>\n",
" <td> 1989-09-09</td>\n",
" <td> NaN</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 2002043480</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> 0</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0003</th>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> Rayyan Jamal Muhyi Al.Deen</td>\n",
" <td> SAra Hussein Muhyi Al.Deen</td>\n",
" <td> 1965-01-01</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 2001860539</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0004</th>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> Hanan Mohd Mustapha Abu Othman</td>\n",
" <td> Kawla Abu Shanab</td>\n",
" <td> 1983-10-31</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 2001953536</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td>...</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0005</th>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> Yara Mahmoud Azmi Ismael</td>\n",
" <td> Suha Abdel Aziz</td>\n",
" <td> 1986-02-28</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 1</td>\n",
" <td> NaN</td>\n",
" <td> 2002040652</td>\n",
" <td> 3</td>\n",
" <td> NaN</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td>...</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows \u00d7 412 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 8,
"text": [
" greater_48hrs fever_neutropenia never_left written_consent \\\n",
"case_id \n",
"A0001 0 0 0 1 \n",
"A0002 0 0 0 1 \n",
"A0003 0 0 0 1 \n",
"A0004 0 0 0 1 \n",
"A0005 0 0 0 1 \n",
"\n",
" child_name mother_name \\\n",
"case_id \n",
"A0001 Remas Mahmoud Jbarah Huda Katalo \n",
"A0002 Majed Abdel Kareem Majed Noor SHa'aban Mahmood \n",
"A0003 Rayyan Jamal Muhyi Al.Deen SAra Hussein Muhyi Al.Deen \n",
"A0004 Hanan Mohd Mustapha Abu Othman Kawla Abu Shanab \n",
"A0005 Yara Mahmoud Azmi Ismael Suha Abdel Aziz \n",
"\n",
" mother_birth_date mother_record mother_nationality \\\n",
"case_id \n",
"A0001 1976-01-21 NaN 3 \n",
"A0002 1989-09-09 NaN 3 \n",
"A0003 1965-01-01 NaN 1 \n",
"A0004 1983-10-31 NaN 1 \n",
"A0005 1986-02-28 NaN 1 \n",
"\n",
" other_mother_nationality father_nationality other_father_nationality \\\n",
"case_id \n",
"A0001 NaN 1 NaN \n",
"A0002 NaN 1 NaN \n",
"A0003 NaN 1 NaN \n",
"A0004 NaN 1 NaN \n",
"A0005 NaN 1 NaN \n",
"\n",
" child_record newborn_id newborn_id_number adm_ari adm_apnea \\\n",
"case_id \n",
"A0001 2002048926 3 NaN 0 0 \n",
"A0002 2002043480 3 NaN 0 0 \n",
"A0003 2001860539 3 NaN 0 0 \n",
"A0004 2001953536 3 NaN 0 0 \n",
"A0005 2002040652 3 NaN 0 0 \n",
"\n",
" adm_asthma adm_bronchiolitis adm_bronchopneumo \n",
"case_id \n",
"A0001 0 0 0 ... \n",
"A0002 0 1 0 ... \n",
"A0003 0 0 0 ... \n",
"A0004 0 0 1 ... \n",
"A0005 0 0 0 ... \n",
"\n",
"[5 rows x 412 columns]"
]
}
],
"prompt_number": 8
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.child_birth_date = pd.to_datetime(hospitalized.child_birth_date)\n",
"hospitalized.enrollment_date = pd.to_datetime(hospitalized.enrollment_date)\n",
"hospitalized.admission_date = pd.to_datetime(hospitalized.admission_date)\n",
"hospitalized.discharge_date = pd.to_datetime(hospitalized.discharge_date)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 9
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized[\"prev_cond\"] = hospitalized[[c for c in hospitalized.columns if c.endswith('hx') and not c.startswith('no_')]].sum(1)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 10
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"[c for c in hospitalized.columns if c.endswith('hx') and not c.startswith('no_')]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 11,
"text": [
"['diabetes_hx',\n",
" 'heart_hx',\n",
" 'kidney_hx',\n",
" 'downs_hx',\n",
" 'sickle_hx',\n",
" 'cancer_hx',\n",
" 'cf_hx',\n",
" 'genetic_hx',\n",
" 'cp_hx',\n",
" 'seizure_hx',\n",
" 'neuro_hx',\n",
" 'mr_hx',\n",
" 'asthma_hx',\n",
" 'immuno_hx',\n",
" 'liver_hx',\n",
" 'gerd_hx',\n",
" 'diarrhea_hx',\n",
" 'other_hx',\n",
" 'spec_hx']"
]
}
],
"prompt_number": 11
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Plot z-scores by year of admission"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['year_admission'] = hospitalized.admission_date.apply(lambda x: x.year)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 12
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.groupby('year_admission')['z_score'].hist(alpha=0.2, bins=20)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 13,
"text": [
"year_admission\n",
"2010 Axes(0.125,0.125;0.775x0.775)\n",
"2011 Axes(0.125,0.125;0.775x0.775)\n",
"2012 Axes(0.125,0.125;0.775x0.775)\n",
"2013 Axes(0.125,0.125;0.775x0.775)\n",
"Name: z_score, dtype: object"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAECCAYAAADw0Rw8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
"<matplotlib.figure.Figure at 0x1108adf50>"
]
}
],
"prompt_number": 13
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Calculate age in months from birth and enrollment date:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from calendar import monthrange\n",
"from datetime import timedelta\n",
"\n",
"def monthdelta(d1, d2):\n",
" delta = 0\n",
" while True:\n",
" mdays = monthrange(d1.year, d1.month)[1]\n",
" d1 += timedelta(days=mdays)\n",
" if d1 <= d2:\n",
" delta += 1\n",
" else:\n",
" break\n",
" return delta"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 14
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"if RSV_only:\n",
" hospitalized = hospitalized[hospitalized.pcr_result___1==1]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 15
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Covariates"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Either ICU or direct admission to ICU"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['icu_any'] = (hospitalized.icu + hospitalized.dir_icu).astype(bool)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 16
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Severity score"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.flaring.value_counts()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 17,
"text": [
"0 1879\n",
"1 1009\n",
"2 264\n",
"3 17\n",
"dtype: int64"
]
}
],
"prompt_number": 17
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.cyanosis[hospitalized.cyanosis==4] = np.nan\n",
"hospitalized.cyanosis.value_counts()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 18,
"text": [
"0 2549\n",
"1 558\n",
"2 51\n",
"3 8\n",
"dtype: int64"
]
}
],
"prompt_number": 18
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.wheezing[hospitalized.wheezing==4] = np.nan\n",
"hospitalized.wheezing.value_counts()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 19,
"text": [
"1 1475\n",
"0 1412\n",
"2 270\n",
"3 10\n",
"dtype: int64"
]
}
],
"prompt_number": 19
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.respiratory_rate = hospitalized.respiratory_rate.replace({'-': np.nan}).astype(float)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 20
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['respiratory_class'] = 0\n",
"hospitalized.respiratory_class[(hospitalized.respiratory_rate>30) & (hospitalized.respiratory_rate<=45)] = 1\n",
"hospitalized.respiratory_class[(hospitalized.respiratory_rate>45) & (hospitalized.respiratory_rate<=60)] = 2\n",
"hospitalized.respiratory_class[hospitalized.respiratory_rate>60] = 3\n",
"hospitalized.respiratory_class.value_counts()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 21,
"text": [
"1 1648\n",
"0 764\n",
"2 636\n",
"3 121\n",
"dtype: int64"
]
}
],
"prompt_number": 21
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['sats_class'] = 0\n",
"hospitalized.sats_class[(hospitalized.sats_number>=90) & (hospitalized.sats_number<95)] = 1\n",
"hospitalized.sats_class[(hospitalized.sats_number>=85) & (hospitalized.sats_number<90)] = 2\n",
"hospitalized.sats_class[hospitalized.sats_number<85] = 3\n",
"hospitalized.sats_class.value_counts()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 22,
"text": [
"0 3020\n",
"1 131\n",
"2 15\n",
"3 3\n",
"dtype: int64"
]
}
],
"prompt_number": 22
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['severity_score'] = hospitalized[['flaring', 'cyanosis', 'wheezing', 'respiratory_class', 'sats_class']].sum(1)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 23
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.severity_score.hist()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 24,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x110f644d0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAECCAYAAADw0Rw8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x110f72b90>"
]
}
],
"prompt_number": 24
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.icu_any[hospitalized.sats_number<85]"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 25,
"text": [
"case_id\n",
"A0226 False\n",
"C2028 False\n",
"D3055 False\n",
"Name: icu_any, dtype: bool"
]
}
],
"prompt_number": 25
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Month of admission"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['admission_month'] = hospitalized.admission_date.apply(lambda x: x.month)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 26
},
{
"cell_type": "code",
"collapsed": true,
"input": [
"covariates = hospitalized[[\"daycare\", \n",
" \"hospitalized_vitamin_d\", \n",
" \"gest_age\", \n",
" \"heart_hx\",\n",
" \"prev_cond\",\n",
" \"cigarette_smokers\",\n",
" \"cigarette_preg\",\n",
" \"sex_child\",\n",
" \"birth_wt_child\",\n",
" \"breastfed\",\n",
" \"severity_score\",\n",
" \"z_score\",\n",
" \"age_months\",\n",
" \"wheezing\",\n",
" \"admission_month\"\n",
" ]]\n"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 27
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"covariates.birth_wt_child.hist(bins=np.sqrt(len(covariates)))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 28,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x111413090>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXwAAAECCAYAAAD9z2x7AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x111410b90>"
]
}
],
"prompt_number": 28
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"covariates.hospitalized_vitamin_d.hist(bins=np.sqrt(len(covariates)))"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 29,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x111413310>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAECCAYAAADw0Rw8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nx6InIlIci56ISHEseiIixf0vsFe5a83pMlIAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x110f7bd90>"
]
}
],
"prompt_number": 29
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"covariates.head()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>daycare</th>\n",
" <th>hospitalized_vitamin_d</th>\n",
" <th>gest_age</th>\n",
" <th>heart_hx</th>\n",
" <th>prev_cond</th>\n",
" <th>cigarette_smokers</th>\n",
" <th>cigarette_preg</th>\n",
" <th>sex_child</th>\n",
" <th>birth_wt_child</th>\n",
" <th>breastfed</th>\n",
" <th>severity_score</th>\n",
" <th>z_score</th>\n",
" <th>age_months</th>\n",
" <th>wheezing</th>\n",
" <th>admission_month</th>\n",
" </tr>\n",
" <tr>\n",
" <th>case_id</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>A0001</th>\n",
" <td> 0</td>\n",
" <td> 3</td>\n",
" <td> 40</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 2.95</td>\n",
" <td> 1</td>\n",
" <td> 6</td>\n",
" <td> 0.68</td>\n",
" <td> 1</td>\n",
" <td> 0</td>\n",
" <td> 3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0002</th>\n",
" <td> 0</td>\n",
" <td> 4</td>\n",
" <td> 40</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> 3.00</td>\n",
" <td> 1</td>\n",
" <td> 4</td>\n",
" <td>-0.64</td>\n",
" <td> 1</td>\n",
" <td> 1</td>\n",
" <td> 3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0003</th>\n",
" <td> 0</td>\n",
" <td> 35</td>\n",
" <td> 40</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 2.00</td>\n",
" <td> 0</td>\n",
" <td> 3</td>\n",
" <td>-7.59</td>\n",
" <td> 11</td>\n",
" <td> 0</td>\n",
" <td> 3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0004</th>\n",
" <td> 0</td>\n",
" <td> 2</td>\n",
" <td> 38</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 2.70</td>\n",
" <td> 1</td>\n",
" <td> 2</td>\n",
" <td>-0.84</td>\n",
" <td> 7</td>\n",
" <td> 1</td>\n",
" <td> 3</td>\n",
" </tr>\n",
" <tr>\n",
" <th>A0005</th>\n",
" <td> 0</td>\n",
" <td> 6</td>\n",
" <td> 39</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 1</td>\n",
" <td> 0</td>\n",
" <td> 0</td>\n",
" <td> 3.00</td>\n",
" <td> 1</td>\n",
" <td> 2</td>\n",
" <td>-1.68</td>\n",
" <td> 2</td>\n",
" <td> 0</td>\n",
" <td> 3</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows \u00d7 15 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 30,
"text": [
" daycare hospitalized_vitamin_d gest_age heart_hx prev_cond \\\n",
"case_id \n",
"A0001 0 3 40 0 0 \n",
"A0002 0 4 40 0 0 \n",
"A0003 0 35 40 0 1 \n",
"A0004 0 2 38 0 0 \n",
"A0005 0 6 39 0 0 \n",
"\n",
" cigarette_smokers cigarette_preg sex_child birth_wt_child \\\n",
"case_id \n",
"A0001 0 0 0 2.95 \n",
"A0002 1 0 1 3.00 \n",
"A0003 0 0 0 2.00 \n",
"A0004 1 0 0 2.70 \n",
"A0005 1 0 0 3.00 \n",
"\n",
" breastfed severity_score z_score age_months wheezing \\\n",
"case_id \n",
"A0001 1 6 0.68 1 0 \n",
"A0002 1 4 -0.64 1 1 \n",
"A0003 0 3 -7.59 11 0 \n",
"A0004 1 2 -0.84 7 1 \n",
"A0005 1 2 -1.68 2 0 \n",
"\n",
" admission_month \n",
"case_id \n",
"A0001 3 \n",
"A0002 3 \n",
"A0003 3 \n",
"A0004 3 \n",
"A0005 3 \n",
"\n",
"[5 rows x 15 columns]"
]
}
],
"prompt_number": 30
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"covariates.daycare.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 31,
"text": [
"0.015782828282828284"
]
}
],
"prompt_number": 31
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"covariates.heart_hx.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 32,
"text": [
"0.046071315872514992"
]
}
],
"prompt_number": 32
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"covariates.cigarette_preg.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 33,
"text": [
"0.076364783843483747"
]
}
],
"prompt_number": 33
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#axes = pd.scatter_matrix(covariates, figsize=(16,12))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 34
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity = hospitalized[[\"oxygen\", \"days_oxygen\", \n",
" \"vent\", \"days_vent\", \n",
" \"icu_any\", \n",
" \"length_of_stay\",\n",
" \"death\"]]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 35
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Some measures of severity are rare"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity.icu_any.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 36,
"text": [
"0.098453770905648469"
]
}
],
"prompt_number": 36
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity.oxygen.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 37,
"text": [
"0.32281708094327599"
]
}
],
"prompt_number": 37
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity.vent.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 38,
"text": [
"0.035384124960153016"
]
}
],
"prompt_number": 38
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Histogram of days on oxygen"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity.days_oxygen.hist(bins=25)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 39,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x1115e0450>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAECCAYAAADw0Rw8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n9ADgcBQ9ADjc/wdR66R5NaiTZQAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x110f77610>"
]
}
],
"prompt_number": 39
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity.death.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 40,
"text": [
"0.0097822656989586618"
]
}
],
"prompt_number": 40
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Attributes of non-survivors"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.age_months[hospitalized.death].hist(bins=25)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 41,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x111bdbcd0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXMAAAECCAYAAAAMxDf2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAMMcADzAMAcAD/w/R8rvq81bqlkAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x111bda290>"
]
}
],
"prompt_number": 41
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.length_of_stay[hospitalized.death].hist(bins=15)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 42,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x111638f50>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAECCAYAAADelD2uAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIGQobgAIGYobAEKG4gaAkPlfQfT2Lx4pITwAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x111c95090>"
]
}
],
"prompt_number": 42
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.gest_age[hospitalized.death].hist(bins=25)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 43,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x111bc6110>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXMAAAECCAYAAAAMxDf2AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDKDMAcCA/wPm3YyBbmUQHgAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x111bd6b50>"
]
}
],
"prompt_number": 43
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized.gest_age<37).mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 44,
"text": [
"0.14200063111391606"
]
}
],
"prompt_number": 44
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized[hospitalized.death].gest_age<37).mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 45,
"text": [
"0.16129032258064516"
]
}
],
"prompt_number": 45
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion positive blood culture"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.blood_neg[hospitalized.blood_neg<2].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 46,
"text": [
"0.12827225130890052"
]
}
],
"prompt_number": 46
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.blood_neg[hospitalized.blood_neg<2][hospitalized.death].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 47,
"text": [
"0.23076923076923078"
]
}
],
"prompt_number": 47
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.blood_pneumo.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 48,
"text": [
"0.021276595744680851"
]
}
],
"prompt_number": 48
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.blood_pneumo[hospitalized.death].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 49,
"text": [
"0.0"
]
}
],
"prompt_number": 49
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Histogram of days on ventilation"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"severity.days_vent.hist(bins=15)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 50,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x111cb60d0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAECCAYAAADuGCyPAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nu1gAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x1122b5dd0>"
]
}
],
"prompt_number": 50
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized[hospitalized.death].vent.median()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 51,
"text": [
"0.0"
]
}
],
"prompt_number": 51
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.death[hospitalized.vent.astype(bool)].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 52,
"text": [
"0.055944055944055944"
]
}
],
"prompt_number": 52
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"outcome = 'oxygen'\n",
"complete = (covariates.isnull().sum(axis=1).astype(bool)==False) & (severity[outcome].isnull().astype(bool)==False)\n",
"covariates_complete = covariates[complete]\n",
"y_complete = severity[outcome][complete]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 67
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables = pd.concat([covariates, severity], axis=1)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 68
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Drop rows not worth imputing"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables[['cigarette_smokers', 'birth_wt_child', \n",
" 'oxygen', 'length_of_stay', 'hospitalized_vitamin_d', 'breastfed']].isnull().sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 69,
"text": [
"cigarette_smokers 1\n",
"birth_wt_child 2\n",
"oxygen 31\n",
"length_of_stay 29\n",
"hospitalized_vitamin_d 480\n",
"breastfed 0\n",
"dtype: int64"
]
}
],
"prompt_number": 69
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables = variables.dropna(subset=['cigarette_smokers', 'birth_wt_child', \n",
" 'oxygen', 'length_of_stay', 'hospitalized_vitamin_d', 'breastfed', 'z_score'])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 70
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Standardize vitamin D, age in months, and birth weight variables"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables['vitamin_d_norm'] = ((variables.hospitalized_vitamin_d - variables.hospitalized_vitamin_d.mean()) \n",
" / variables.hospitalized_vitamin_d.std())\n",
"\n",
"variables['wt_norm'] = ((variables.birth_wt_child - variables.birth_wt_child.mean()) /\n",
" variables.birth_wt_child.std())\n",
"\n",
"variables['age_norm'] = ((variables.age_months - variables.age_months.mean()) /\n",
" variables.age_months.std())"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 71
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Convert gestational age to *number of weeks less than 37*."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables['premature'] = np.maximum(0, 37 - variables.gest_age)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 72
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables.premature.hist(bins=20, grid=False)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 73,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x11242f950>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAECCAYAAAD3vwBsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nfJx9hU4CAAAAAElFTkSuQmCC\n",
"text": [
"<matplotlib.figure.Figure at 0x1140b8dd0>"
]
}
],
"prompt_number": 73
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Extract usable subset of covariates and response"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables_glm = variables[\n",
" ['vitamin_d_norm', 'premature', 'cigarette_smokers', 'sex_child', \n",
" 'wt_norm', 'oxygen', 'breastfed', 'z_score', 'age_norm']]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 74
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Confirm no nulls"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables_glm.isnull().sum(0)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 75,
"text": [
"vitamin_d_norm 0\n",
"premature 0\n",
"cigarette_smokers 0\n",
"sex_child 0\n",
"wt_norm 0\n",
"oxygen 0\n",
"breastfed 0\n",
"z_score 0\n",
"age_norm 0\n",
"dtype: int64"
]
}
],
"prompt_number": 75
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"data_glm = variables_glm.to_dict(outtype='list')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 76
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from pymc import Model, glm\n",
"import theano.tensor as t\n",
"from pymc import sample, Slice, NUTS\n",
"from pymc import forestplot, traceplot, summary\n",
"\n",
"def tinvlogit(x):\n",
" return t.exp(x) / (1 + t.exp(x))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 58
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"with Model() as interactive_model:\n",
" \n",
" formula = 'oxygen ~ (vitamin_d_norm + premature + cigarette_smokers + sex_child '\n",
" formula += '+ breastfed + z_score + age_norm) ** 2'\n",
" \n",
" glm.glm(formula, data_glm,\n",
" family=glm.families.Binomial(link=glm.links.Logit))"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 59
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"with interactive_model:\n",
" trace_int = sample(2000, NUTS(interactive_model.vars), progressbar=False) "
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stderr",
"text": [
"/Library/Python/2.7/site-packages/theano/scan_module/scan_perform_ext.py:85: RuntimeWarning: numpy.ndarray size changed, may indicate binary incompatibility\n",
" from scan_perform.scan_perform import *\n"
]
},
{
"ename": "KeyboardInterrupt",
"evalue": "",
"output_type": "pyerr",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m\n\u001b[0;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-60-1a5994b5a2ed>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m 1\u001b[0m \u001b[0;32mwith\u001b[0m \u001b[0minteractive_model\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0mtrace_int\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0msample\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;36m2000\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mNUTS\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minteractive_model\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvars\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mprogressbar\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;32m/Users/fonnescj/Code/pymc/pymc/step_methods/nuts.pyc\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, vars, scaling, step_scale, is_cov, state, Emax, target_accept, gamma, k, t0, model)\u001b[0m\n\u001b[1;32m 57\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 58\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mscaling\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdict\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 59\u001b[0;31m \u001b[0mscaling\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mguess_scaling\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mPoint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mscaling\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvars\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 60\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 61\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Code/pymc/pymc/tuning/scaling.pyc\u001b[0m in \u001b[0;36mguess_scaling\u001b[0;34m(point, vars, model)\u001b[0m\n\u001b[1;32m 77\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mguess_scaling\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 78\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodelcontext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 79\u001b[0;31m \u001b[0mh\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfind_hessian_diag\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 80\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0madjust_scaling\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mh\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 81\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Code/pymc/pymc/tuning/scaling.pyc\u001b[0m in \u001b[0;36mfind_hessian_diag\u001b[0;34m(point, vars, model)\u001b[0m\n\u001b[1;32m 72\u001b[0m \"\"\"\n\u001b[1;32m 73\u001b[0m \u001b[0mmodel\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodelcontext\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 74\u001b[0;31m \u001b[0mH\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfastfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhessian_diag\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlogpt\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 75\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mH\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mPoint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mpoint\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmodel\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 76\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Code/pymc/pymc/memoize.pyc\u001b[0m in \u001b[0;36mmemoizer\u001b[0;34m(*args, **kwargs)\u001b[0m\n\u001b[1;32m 12\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 13\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mkey\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mcache\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 14\u001b[0;31m \u001b[0mcache\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mobj\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 15\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 16\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mcache\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mkey\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Code/pymc/pymc/theanof.pyc\u001b[0m in \u001b[0;36mhessian_diag\u001b[0;34m(f, vars)\u001b[0m\n\u001b[1;32m 92\u001b[0m \u001b[0mvars\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mcont_inputs\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 93\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 94\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0;34m-\u001b[0m\u001b[0mt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconcatenate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mhessian_diag1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 95\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 96\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Code/pymc/pymc/theanof.pyc\u001b[0m in \u001b[0;36mhessian_diag1\u001b[0;34m(f, v)\u001b[0m\n\u001b[1;32m 84\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mgradient1\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mg\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mv\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 85\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 86\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmap\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mhess_ii\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0midx\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 87\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 88\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/scan_module/scan_views.pyc\u001b[0m in \u001b[0;36mmap\u001b[0;34m(fn, sequences, non_sequences, truncate_gradient, go_backwards, mode, name)\u001b[0m\n\u001b[1;32m 66\u001b[0m \u001b[0mgo_backwards\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mgo_backwards\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 67\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mmode\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 68\u001b[0;31m name=name)\n\u001b[0m\u001b[1;32m 69\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 70\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/scan_module/scan.pyc\u001b[0m in \u001b[0;36mscan\u001b[0;34m(fn, sequences, outputs_info, non_sequences, n_steps, truncate_gradient, go_backwards, mode, name, profile)\u001b[0m\n\u001b[1;32m 1005\u001b[0m \u001b[0;32mpass\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1006\u001b[0m \u001b[0mscan_inputs\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0marg\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1007\u001b[0;31m \u001b[0mscan_outs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mlocal_op\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m*\u001b[0m\u001b[0mscan_inputs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1008\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mtype\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mscan_outs\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mlist\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtuple\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1009\u001b[0m \u001b[0mscan_outs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mscan_outs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/op.pyc\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *inputs, **kwargs)\u001b[0m\n\u001b[1;32m 458\u001b[0m \u001b[0;31m# compute output value once with test inputs to validate graph\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 459\u001b[0m thunk = node.op.make_thunk(node, storage_map, compute_map,\n\u001b[0;32m--> 460\u001b[0;31m no_recycling=[])\n\u001b[0m\u001b[1;32m 461\u001b[0m \u001b[0mthunk\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mstorage_map\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 462\u001b[0m \u001b[0mthunk\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0mstorage_map\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mv\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mv\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/scan_module/scan_op.pyc\u001b[0m in \u001b[0;36mmake_thunk\u001b[0;34m(self, node, storage_map, compute_map, no_recycling)\u001b[0m\n\u001b[1;32m 577\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 578\u001b[0m \u001b[0mprofile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 579\u001b[0;31m on_unused_input='ignore')\n\u001b[0m\u001b[1;32m 580\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 581\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/compile/function.pyc\u001b[0m in \u001b[0;36mfunction\u001b[0;34m(inputs, outputs, mode, updates, givens, no_default_updates, accept_inplace, name, rebuild_strict, allow_input_downcast, profile, on_unused_input)\u001b[0m\n\u001b[1;32m 221\u001b[0m \u001b[0mallow_input_downcast\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mallow_input_downcast\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 222\u001b[0m \u001b[0mon_unused_input\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mon_unused_input\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 223\u001b[0;31m profile=profile)\n\u001b[0m\u001b[1;32m 224\u001b[0m \u001b[0;31m# We need to add the flag check_aliased inputs if we have any mutable or\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 225\u001b[0m \u001b[0;31m# borrowed used defined inputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/compile/pfunc.pyc\u001b[0m in \u001b[0;36mpfunc\u001b[0;34m(params, outputs, mode, updates, givens, no_default_updates, accept_inplace, name, rebuild_strict, allow_input_downcast, profile, on_unused_input)\u001b[0m\n\u001b[1;32m 510\u001b[0m return orig_function(inputs, cloned_outputs, mode,\n\u001b[1;32m 511\u001b[0m \u001b[0maccept_inplace\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maccept_inplace\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mname\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mprofile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 512\u001b[0;31m on_unused_input=on_unused_input)\n\u001b[0m\u001b[1;32m 513\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 514\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/compile/function_module.pyc\u001b[0m in \u001b[0;36morig_function\u001b[0;34m(inputs, outputs, mode, accept_inplace, name, profile, on_unused_input)\u001b[0m\n\u001b[1;32m 1325\u001b[0m \u001b[0maccept_inplace\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0maccept_inplace\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1326\u001b[0m \u001b[0mprofile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mprofile\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1327\u001b[0;31m \u001b[0mon_unused_input\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mon_unused_input\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcreate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1328\u001b[0m defaults)\n\u001b[1;32m 1329\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/compile/function_module.pyc\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, inputs, outputs, mode, accept_inplace, function_builder, profile, on_unused_input)\u001b[0m\n\u001b[1;32m 1036\u001b[0m \u001b[0mgof\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mOp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0madd_stack_trace_on_call\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1037\u001b[0m \u001b[0mstart_optimizer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1038\u001b[0;31m \u001b[0moptimizer_profile\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1039\u001b[0m \u001b[0mend_optimizer\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1040\u001b[0m \u001b[0mopt_time\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mend_optimizer\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mstart_optimizer\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/opt.pyc\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, fgraph)\u001b[0m\n\u001b[1;32m 89\u001b[0m \u001b[0mSame\u001b[0m \u001b[0;32mas\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 90\u001b[0m \"\"\"\n\u001b[0;32m---> 91\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 92\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 93\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0madd_requirements\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/opt.pyc\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, fgraph, *args, **kwargs)\u001b[0m\n\u001b[1;32m 80\u001b[0m \u001b[0morig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 81\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 82\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 83\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 84\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0morig\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/opt.pyc\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, fgraph)\u001b[0m\n\u001b[1;32m 194\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 195\u001b[0m \u001b[0mt0\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 196\u001b[0;31m \u001b[0msub_prof\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptimize\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 197\u001b[0m \u001b[0ml\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfloat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 198\u001b[0m \u001b[0msub_profs\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mappend\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msub_prof\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/opt.pyc\u001b[0m in \u001b[0;36moptimize\u001b[0;34m(self, fgraph, *args, **kwargs)\u001b[0m\n\u001b[1;32m 80\u001b[0m \u001b[0morig\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 81\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 82\u001b[0;31m \u001b[0mret\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mapply\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 83\u001b[0m \u001b[0;32mfinally\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 84\u001b[0m \u001b[0mtheano\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbasic\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mconstant\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0menable\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0morig\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/opt.pyc\u001b[0m in \u001b[0;36mapply\u001b[0;34m(self, fgraph, start_from)\u001b[0m\n\u001b[1;32m 1652\u001b[0m self.local_optimizers_map.get(node.op, [])):\n\u001b[1;32m 1653\u001b[0m \u001b[0mt_opt\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1654\u001b[0;31m \u001b[0mlopt_change\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mprocess_node\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfgraph\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mlopt\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1655\u001b[0m \u001b[0mtime_opts\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlopt\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt_opt\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1656\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlopt_change\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/opt.pyc\u001b[0m in \u001b[0;36mprocess_node\u001b[0;34m(self, fgraph, node, lopt)\u001b[0m\n\u001b[1;32m 1350\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mFalse\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1351\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1352\u001b[0;31m \u001b[0mfgraph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreplace_all_validate\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrepl_pairs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreason\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlopt\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1353\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1354\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/toolbox.pyc\u001b[0m in \u001b[0;36mreplace_all_validate\u001b[0;34m(self, fgraph, replacements, reason, verbose)\u001b[0m\n\u001b[1;32m 206\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnew_r\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mreplacements\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 207\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 208\u001b[0;31m \u001b[0mfgraph\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mreplace\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnew_r\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreason\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mreason\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mFalse\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 209\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mException\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0me\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 210\u001b[0m if ('The type of the replacement must be the same' not in\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/fg.pyc\u001b[0m in \u001b[0;36mreplace\u001b[0;34m(self, r, new_r, reason, verbose)\u001b[0m\n\u001b[1;32m 472\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mlist\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mclients\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;31m# copy the client list for iteration\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 473\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m'output'\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moutputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minputs\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 474\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mchange_input\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnew_r\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mreason\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mreason\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 475\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 476\u001b[0m \u001b[0;31m# sometimes the following is triggered. If you understand why, please explain to James.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/fg.pyc\u001b[0m in \u001b[0;36mchange_input\u001b[0;34m(self, node, i, new_r, reason)\u001b[0m\n\u001b[1;32m 429\u001b[0m \u001b[0;31m# transaction will be reverted later.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 430\u001b[0m self.execute_callbacks('on_change_input', node, i,\n\u001b[0;32m--> 431\u001b[0;31m r, new_r, reason=reason)\n\u001b[0m\u001b[1;32m 432\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 433\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mprune\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/gof/fg.pyc\u001b[0m in \u001b[0;36mexecute_callbacks\u001b[0;34m(self, name, *args, **kwargs)\u001b[0m\n\u001b[1;32m 552\u001b[0m \u001b[0;32mcontinue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 553\u001b[0m \u001b[0mtf0\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 554\u001b[0;31m \u001b[0mfn\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m*\u001b[0m\u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwargs\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 555\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexecute_callbacks_times\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mfeature\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mtf0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 556\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mexecute_callbacks_time\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0mtime\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtime\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mt0\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/tensor/opt.pyc\u001b[0m in \u001b[0;36mon_change_input\u001b[0;34m(self, fgraph, node, i, r, new_r, reason)\u001b[0m\n\u001b[1;32m 1065\u001b[0m \u001b[0;31m# owner(i.e. it is a constant or an input of the graph)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1066\u001b[0m \u001b[0;31m# update_shape suppose that r and new_r are in shape_of.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1067\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0minit_r\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mnew_r\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1068\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1069\u001b[0m \u001b[0;31m# This tells us that r and new_r must have the same shape if\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/tensor/opt.pyc\u001b[0m in \u001b[0;36minit_r\u001b[0;34m(self, r)\u001b[0m\n\u001b[1;32m 946\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mr\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape_of\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 947\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 948\u001b[0;31m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_shape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape_tuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 949\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mAttributeError\u001b[0m\u001b[0;34m:\u001b[0m \u001b[0;31m# XXX: where would this come from?\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 950\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mset_shape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mNone\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/tensor/opt.pyc\u001b[0m in \u001b[0;36mshape_tuple\u001b[0;34m(self, r)\u001b[0m\n\u001b[1;32m 754\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mshape_tuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 755\u001b[0m \u001b[0;34m\"\"\"Return a tuple of symbolic shape vars for tensor variable r\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 756\u001b[0;31m \u001b[0;32mreturn\u001b[0m \u001b[0mtuple\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mshape_ir\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mi\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mxrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mndim\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 757\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 758\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mdefault_infer_shape\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mnode\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi_shapes\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Library/Python/2.7/site-packages/theano/tensor/opt.pyc\u001b[0m in \u001b[0;36mshape_ir\u001b[0;34m(self, i, r)\u001b[0m\n\u001b[1;32m 741\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mshape_ir\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mi\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 742\u001b[0m \u001b[0;34m\"\"\"Return symbolic r.shape[i] for tensor variable r, int i\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 743\u001b[0;31m \u001b[0;32mif\u001b[0m \u001b[0mhasattr\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtype\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m\"broadcastable\"\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mand\u001b[0m \u001b[0mr\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mtype\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbroadcastable\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mi\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 744\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlscalar_one\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 745\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"prompt_number": 60
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#_ = traceplot(trace_int)"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Forest plot of parameter estimates. Thick line is the interquartile range of the estimate, thin line the 95% posterior interval. Positive values mean an *increase* in the probability of ending up on oxygen when the value of that parameter increasees by one unit.\n",
"\n",
"So, for example, the z-score main effect is negative, so it decreases the probability of oxygen when it increases (which makes sense). The interaction of z-score with normalized age (bottom of plot) is negative, meaning that individuals that are older with higher z-scores even further reduce the probability of being on oxygen (non-additive effect)."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.figure(figsize=(14,8))\n",
"forestplot(trace_int, vars=trace_int.varnames[1:])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"variables_rsv = variables.ix[hospitalized.pcr_result___1==1,\n",
" ['vitamin_d_norm', 'premature', 'cigarette_smokers', 'sex_child', 'wt_norm', \n",
" 'oxygen', 'breastfed', 'z_score', 'age_norm']].to_dict(outtype='list')"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"with Model() as rsv_model:\n",
" \n",
" formula = 'oxygen ~ (vitamin_d_norm + premature + cigarette_smokers + sex_child + breastfed + z_score + age_norm) ** 2'\n",
" \n",
" glm.glm(formula, variables_rsv,\n",
" family=glm.families.Binomial(link=glm.links.Logit))"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"with rsv_model:\n",
" trace_rsv = sample(2000, NUTS(), progressbar=False) "
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"plt.figure(figsize=(14,8))\n",
"forestplot(trace_rsv, vars=trace_rsv.varnames[1:])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Additional analyses"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Distribution of smoking exposure"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.cigarette_smokers.hist()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 77,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x114102850>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAECCAYAAAD3vwBsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n/T+u8R/veML5RQAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x1140c4bd0>"
]
}
],
"prompt_number": 77
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion and of children exposed to cigarette smoking"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized.cigarette_smokers > 0).mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 78,
"text": [
"0.72578100347112651"
]
}
],
"prompt_number": 78
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized.cigarette_smokers > 0).sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 79,
"text": [
"2300"
]
}
],
"prompt_number": 79
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Distribution of nargila exposure"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.nargila_smokers.hist()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 80,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x112366890>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAX0AAAECCAYAAAASDQdFAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nX0TEQ/4//Eu4cz3nvrUAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x11410fe10>"
]
}
],
"prompt_number": 80
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion and number of children exposed to nargila"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized.nargila_smokers > 0).mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 81,
"text": [
"0.17828968128747238"
]
}
],
"prompt_number": 81
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized.nargila_smokers > 0).sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 82,
"text": [
"565"
]
}
],
"prompt_number": 82
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion of children with mothers who smoked during pregnancy"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.cigarette_preg.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 83,
"text": [
"0.076364783843483747"
]
}
],
"prompt_number": 83
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.nargila_preg.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 84,
"text": [
"0.028400126222783213"
]
}
],
"prompt_number": 84
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Days of symptoms before admission by gender"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"days_by_sex = hospitalized.groupby('sex_child')\n",
"_ = days_by_sex.boxplot(column='days_symptoms')\n"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAX8AAAEHCAYAAABGNUbLAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nUviLiBhI4S8iYiCFv4iIgRT+IiIG+v83T/xoCBtR+wAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x11439bdd0>"
]
}
],
"prompt_number": 85
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Vitamin D levels by breastfeeding status"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"TODO: clean up plot for presentation"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.breastfed = hospitalized.breastfed.replace({0: 'Not breastfed', 1: 'Breastfed'})\n",
"# _ = hospitalized.groupby('breastfed').boxplot(column='hospitalized_vitamin_d', grid=False, fontsize=0)\n",
"# plt.gca().set_ylabel('Vitamin D levels')\n",
"for index, group in hospitalized.groupby(['breastfed']):\n",
" group.boxplot(column='hospitalized_vitamin_d')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAD8CAYAAACSCdTiAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAEyMkAcAEyPkAcDECHkAMDFCHgBM7P8Ba/W1fJ/P89gAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x111c8c6d0>"
]
}
],
"prompt_number": 86
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"_ = hospitalized.groupby('breastfed')['oxygen'].hist(alpha=0.3)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAECCAYAAAAYfWtSAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAHWpJREFUeJzt3X1MXGXe//E3BSstSscZpewuse4KlWpcGwvqhsZCm91N\nZ92IIZJUTXZxjYloVxQWTKOVxtCHLI00FTQ+RN0HEzepEneDu648VKgmhVJW/KECdavlvouzZaZU\n2gIF5veHWy65C5SZU5hDz+eVNOE6M8O5zifT+fZ8rzmnUcFgMIiIiDjSgkhPQEREIkdFQETEwVQE\nREQcTEVARMTBVARERBxMRUBExMFipnuwqqqKgwcPEh8fz86dOyc89te//pU//elPvPLKK1x22WUA\n1NTUUFdXR3R0NHl5eaSmpgLQ09NDVVUVw8PDrFq1ig0bNszS4YiISCimPRPIyspi06ZN52w/duwY\nH3/8MVdeeeX4tp6eHurr69m+fTuFhYVUVlZy9hKEyspK7r//fsrLyzl8+DBtbW0X+DBERCQc0xaB\nFStWEBcXd872P/zhD9x3330TtjU3N5ORkUFMTAwJCQkkJibS1dVFIBBgcHCQ5ORkAG6//Xb2799/\nAQ9BRETCFfKaQHNzM263m2XLlk3YHggE8Hg842OPx4Pf7ycQCOB2u8e3u91u/H6/hSmLiMiFElIR\nGBoa4u233yY3N3d8m+46ISIyf027MPx/ff311/znP//hd7/7HQB+v58nnniCsrIy3G43fX1948/t\n6+vD4/Gc8y//vr6+CWcGk6mtrQ1lWiIi8l/r1q0L6fkhFYGrr76al156aXz88MMPs2PHDi677DLS\n0tLYtWsXd9xxB36/n97eXpKTk4mKimLRokV0dXWRnJxMY2Mj69evP+++br755pAORETE6VpbW0N+\nzbTtoIqKCp566imOHj3KQw89RH19/YTHo6Kixn9OSkoiKyuLkpISysvLyc/PH388Pz+fV199laKi\nIpYtW8bKlStDnqhTNTU1RXoKtqEsDGVhKAtrpj0TKCgomPbFzz333ISx1+vF6/We87ykpCS2bt0a\nxvRERGQ2Rdnx/xOora1VO0hEJEStra0hrwnothEiIg6mImBz6ncaysJQFoaysEZFQETEwbQmICJy\nkdCagIiIhERFwObU7zSUhaEsDGVhjYqAiIiDaU1AROQiEc6aQEj3DppL/9s/FNH9uxfHEHtJdETn\nICIy22xbBL48PhjR/cctXGyLItDU1MTq1asjPQ1bUBaGsjCUhTVaExARcTDbrgkMea6N6ByuT1jM\nkkWXRHQOIiKh0HUCIiISEhUBm9N3oA1lYSgLQ1lYoyIgIuJgWhOYgtYERGS+0ZqAiIiEREXA5tTv\nNJSFoSwMZWGNioCIiINpTWAKWhMQkflGawIiIhISFQGbU7/TUBaGsjCUhTXT3kCuqqqKgwcPEh8f\nz86dOwH44x//SGtrKwsXLmTFihXk5uayePFiAGpqaqirqyM6Opq8vDxSU1MB6OnpoaqqiuHhYVat\nWsWGDRtm+bBERGQmpj0TyMrKYtOmTRO23XTTTezcuZNt27YxODjI22+/DXz7QV9fX8/27dspLCyk\nsrKSs8sNlZWV3H///ZSXl3P48GHa2tpm6XAuPro7oqEsDGVhKAtrpi0CK1asIC4ubsK2H//4xyxY\nsIAFCxawcuVK+vr6AGhubiYjI4OYmBgSEhJITEykq6uLQCDA4OAgycnJANx+++3s379/lg5HRERC\nYWlNoLa2lvT0dAACgQAej2f8MY/Hg9/vJxAI4Ha7x7e73W78fr+V3TqK+p2GsjCUhaEsrAm7CLz1\n1lvExsbyk5/85ELOZ1xHR8eEn+d6/K+P/zU+bmpqmvBG01jjSI/b29ttNZ9Ijtvb2201n0iPQ3Xe\n6wR8Ph87duwYXxgGaGhooLa2lqeeeoqFCxcCUF1dDUB2djYAZWVl5ObmctVVV7FlyxaeffbZ8Ql3\ndHTw4IMPTrlPXScgIhK6OblOoK2tjXfeeYfi4uLxAgCQlpbGvn37GBkZwefz0dvbS3JyMi6Xi0WL\nFtHV1UUwGKSxsZFbbrkl1N2KiMgsmPZMoKKigk8//ZQTJ07gcrm4++67qa6uZmRkhMsuuwyA5cuX\n88ADDwDffkW0trZ2/CuiK1asAMxXRIeGhli1ahX33HPPtJPSmYDR1KT/P/UsZWEoC0NZGOGcCUx7\nnUBBQcE529auXTvl871eL16v95ztSUlJbN26NaSJiYjI7NO9g6ZglzMBEZGZ0r2DREQkJCoCNmfl\nq18XG2VhKAtDWVijIiAi4mBaE5iC1gREZL7RmoCIiIRERcDm1O80lIWhLAxlYY2KgIiIg2lNYApa\nExCR+UZrAiIiEhIVAZtTv9NQFoayMJSFNSoCIiIOpjWBKWhNQETmG60JiIhISFQEbE79TkNZGMrC\nUBbWqAiIiDiY1gSmoDUBEZlvtCYgIiIhURGwOfU7DWVhKAtDWVijIiAi4mBaE5iC1gREZL7RmoCI\niIRERcDm1O80lIWhLAxlYU3MdA9WVVVx8OBB4uPj2blzJwCnT59m9+7d+Hw+li5dysaNG4mNjQWg\npqaGuro6oqOjycvLIzU1FYCenh6qqqoYHh5m1apVbNiwYZYPS0REZmLaM4GsrCw2bdo0YduePXu4\n7rrrKC8vJyUlhT179gDfftDX19ezfft2CgsLqays5OxyQ2VlJffffz/l5eUcPnyYtra2WTqci8/q\n1asjPQXbUBaGsjCUhTXTFoEVK1YQFxc3YVtLSwtr1qwBIDMzk+bmZgCam5vJyMggJiaGhIQEEhMT\n6erqIhAIMDg4SHJyMgC33347+/fvn41jERGREIW8JtDf34/L5QJgyZIl9Pf3AxAIBPB4POPP83g8\n+P1+AoEAbrd7fLvb7cbv91udt2Oo32koC0NZGMrCGksLw1FRURdqHufo6OiY8PNcj//18b/Gx01N\nTRPeaBprHOlxe3u7reYTyXF7e7ut5hPpcajOe52Az+djx44d4wvDBQUFlJaW4nK5CAQCbNmyhYqK\nCqqrqwHIzs4GoKysjNzcXK666iq2bNnCs88+Oz7hjo4OHnzwwSn3qesERERCNyfXCaSlpdHQ0ADA\n3r17SU9PH9++b98+RkZG8Pl89Pb2kpycjMvlYtGiRXR1dREMBmlsbOSWW24JdbciIjILpi0CFRUV\nPPXUUxw9epSHHnqI+vp6cnJy6OzspKioiK6uLnJycgBISkoiKyuLkpISysvLyc/PH28X5efn8+qr\nr1JUVMSyZctYuXLl7B/ZRcLKad7FRlkYysJQFtZMe51AQUHBpNuLi4sn3e71evF6vedsT0pKYuvW\nrWFMT0REZpPuHTQFrQmIyHyjeweJiEhIVARsTv1OQ1kYysJQFtaoCIiIOJjWBKagNQERmW+0JiAi\nIiFREbA59TsNZWEoC0NZWKMiICLiYFoTmILWBERkvtGagIiIhERFwObU7zSUhaEsDGVhjYqAiIiD\naU1gCloTEJH5RmsCIiISEhUBm1O/01AWhrIwlIU1KgIiIg6mNYEpaE1AROYbrQmIiEhIVARsTv1O\nQ1kYysJQFtaoCIiIOJjWBKagNQERmW+0JiAiIiFREbA59TsNZWEoC0NZWBMT7gvff/99GhoaOHPm\nDCtWrODXv/41p0+fZvfu3fh8PpYuXcrGjRuJjY0FoKamhrq6OqKjo8nLyyM1NfWCHYSIiIQnrDOB\ngYEB3n77bZ588km2bdvG0aNHaWtrY8+ePVx33XWUl5eTkpLCnj17AOjp6aG+vp7t27dTWFhIZWUl\nY2NjF/RALlarV6+O9BRsQ1kYysJQFtaEVQQWLlwIwKlTpxgeHmZoaIi4uDhaWlpYs2YNAJmZmTQ3\nNwPQ3NxMRkYGMTExJCQkkJiYSHd397T7GDsxGNE/o8Oj4UQjIjKvhNUOWrhwIQ888AAPP/wwl1xy\nCevXryclJYX+/n5cLhcAS5Ysob+/H4BAIEBKSsr46z0eD36/f9p9HPefCmdqF8zIDy6L6P7Pampq\n0r90/ktZGMrCUBbWhHUmcOLECV5++WWeffZZKisr6ezs5MCBAxOeExUVNe3vON/jXx05MuHnuR53\ndx8aHzc1NU1YfNJY40iP29vbbTWfSI7b29ttNZ9Ij0MV1nUCra2tfPDBBxQUFADw3nvv4fP5aGlp\nobS0FJfLRSAQYMuWLVRUVFBdXQ1AdnY2AGVlZeTm5k44O/iu2tpajn5zRbjHdEGk35hAgjsuonMQ\nEQnFnF0nkJqayqFDhxgYGODMmTMcPHiQm266ibS0NBoaGgDYu3cv6enpAKSlpbFv3z5GRkbw+Xz0\n9vaSnJwczq5FROQCCqsILF68mJycHH7/+9+zefNmli1bxg033EBOTg6dnZ0UFRXR1dVFTk4OAElJ\nSWRlZVFSUkJ5eTn5+fnnbQfJt6yc5l1slIWhLAxlYU3Y1wlkZmaSmZk5YduiRYsoLi6e9Plerxev\n1xvu7kREZBboimGb07ceDGVhKAtDWVijIiAi4mAqAjanfqehLAxlYSgLa1QEREQcTEXA5tTvNJSF\noSwMZWGNioCIiIOpCNic+p2GsjCUhaEsrFEREBFxMBUBm1O/01AWhrIwlIU1KgIiIg6mImBz6nca\nysJQFoaysEZFQETEwVQEbE79TkNZGMrCUBbWqAiIiDiYioDNqd9pKAtDWRjKwhoVARERB1MRsDn1\nOw1lYSgLQ1lYoyIgIuJgKgI2p36noSwMZWEoC2tUBEREHExFwObU7zSUhaEsDGVhjYqAiIiDqQjY\nnPqdhrIwlIWhLKyJCfeFg4ODvPzyy3z11VecOXOG/Px8kpKS2L17Nz6fj6VLl7Jx40ZiY2MBqKmp\noa6ujujoaPLy8khNTb1gByEiIuGJCgaDwXBe+Nxzz3H99dezdu1aRkdHGRoa4q233uLyyy/nzjvv\npLq6mpMnT3LvvffS09PDrl272LZtG36/n2eeeYZdu3axYMHkJyK1tbUc/eYKSwdmVfqNCSS44yI6\nBxGRULS2trJu3bqQXhNWO+jUqVN89tlnrF27FoDo6GgWL15MS0sLa9asASAzM5Pm5mYAmpubycjI\nICYmhoSEBBITE+nu7g5n1yIicgGFVQR8Ph/x8fFUVlZSWFjICy+8wNDQEP39/bhcLgCWLFlCf38/\nAIFAAI/HM/56j8eD3++/ANO/+KnfaSgLQ1kYysKasIrA6Ogohw4d4tZbb2Xbtm2MjIzw0UcfTXhO\nVFTUtL/jfI9/deTIhJ/netzdfWh83NTUNOGNprHGkR63t7fbaj6RHLe3t9tqPpEehyqsNYHjx49T\nWFjIK6+8AsDBgwfZu3cvX375JU8//TQul4tAIMCWLVuoqKiguroagOzsbADKysrIzc0lJSVl0t+v\nNQERmU8GhkYYGB6N9DT4367/F/KaQFjfDnK5XCQmJtLV1cW1115La2srN954I1deeSUNDQ1kZ2ez\nd+9e0tPTAUhLS2PXrl3ccccd+P1+ent7SU5ODmfXIiK2MzA8yr/9g5GeBpeG8ZqwrxN4+OGHee21\n13j00Ufx+/1kZGSQk5NDZ2cnRUVFdHV1kZOTA0BSUhJZWVmUlJRQXl5Ofn7+edtB8i0rp3kXG2Vh\nKAtDWVgT9nUC3//+9ykrKztne3Fx8aTP93q9eL3ecHcnIiKzQFcM25zui2IoC0NZGMrCGhUBEREH\nUxGwOfU7DWVhKAtDWVijIiAi4mAqAjanfqehLAxlYSgLa1QEREQcTEXA5tTvNJSFoSwMZWGNioCI\niIOpCNic+p2GsjCUhaEsrFEREBFxMBUBm1O/01AWhrIwlIU1KgIiIg6mImBz6ncaysJQFoaysEZF\nQETEwVQEbE79TkNZGMrCUBbWqAiIiDiYioDNqd9pKAtDWRjKwhoVARERB1MRsDn1Ow1lYSgLQ1lY\noyIgIuJgKgI2p36noSwMZWEoC2tUBEREHExFwObU7zSUhaEsDGVhTYyVF4+NjfHEE0/gdrt54okn\nOH36NLt378bn87F06VI2btxIbGwsADU1NdTV1REdHU1eXh6pqakX5ABERCR8ls4EampqSEpKIioq\nCoA9e/Zw3XXXUV5eTkpKCnv27AGgp6eH+vp6tm/fTmFhIZWVlYyNjVmfvQOo32koC0NZGHbIYnRo\nBE4NR/5PGMI+E+jr6+PgwYPcdddd/O1vfwOgpaWF0tJSADIzMyktLeXee++lubmZjIwMYmJiSEhI\nIDExke7ubpYvXx7u7kVEbGN4cAS/byDS0+B7l4f+mrDPBF5//XXuu+8+Fiwwv6K/vx+XywXAkiVL\n6O/vByAQCODxeMaf5/F48Pv94e7aUdTvNJSFoSwMZWFNWEXgwIEDxMfH88Mf/pBgMDjpc862iKZy\nvse/OnJkws9zPe7uPjQ+bmpqmvBG01jjSI/b29ttNZ9Ijtvb2yM+n8NfHh4fR+Lz6rvjUEUFp/oU\nn8Ybb7xBY2MjCxYs4MyZM5w+fZpbbrmFQ4cOUVpaisvlIhAIsGXLFioqKqiurgYgOzsbgLKyMnJz\nc0lJSZn099fW1nL0myvCPqgLIf3GBBLccRGdg4jMD//u6afjUOS7G9+7PMC6detCek1YZwL33HMP\nzz//PJWVlRQUFHDDDTewceNG0tLSaGhoAGDv3r2kp6cDkJaWxr59+xgZGcHn89Hb20tycnI4uxYR\nkQvoglwncLa1k5OTQ2dnJ0VFRXR1dZGTkwNAUlISWVlZlJSUUF5eTn5+/nnbQfKt7552Op2yMJSF\noSyssXSdAMD111/P9ddfD8CiRYsoLi6e9Hlerxev12t1dyIicgHpimGbs8N3oO1CWRjKwlAW1qgI\niIg4mIqAzanfaSgLQ1kYysIaFQEREQdTEbA59TsNZWEoC0NZWKMiICLiYCoCNqd+p6EsDGVhKAtr\nVARERBxMRcDm1O80lIWhLAxlYY2KgIiIg6kI2Jz6nYayMJSFoSysUREQEXEwFQGbU7/TUBaGsjCU\nhTUqAiIiDqYiYHPqdxrKwlAWhrKwRkVARMTBVARsTv1OQ1kYysJQFtaoCIiIOJiKgM2p32koC0NZ\nGMrCGhUBEREHUxGwOfU7DWVhKAtDWVijIiAi4mAqAjanfqehLAxlYSgLa2LCedGxY8eorKykv7+f\n+Ph4MjMzyczM5PTp0+zevRufz8fSpUvZuHEjsbGxANTU1FBXV0d0dDR5eXmkpqZe0AMREZHQRQWD\nwWCoLzp+/DjHjx/nmmuu4cSJExQWFvL000/T0NDA5Zdfzp133kl1dTUnT57k3nvvpaenh127drFt\n2zb8fj/PPPMMu3btYsGCyU9EamtrOfrNFZYPzor0GxNIcMdFdA4iMj/8u6efjkP+SE+D710eYN26\ndSG9Jqx2kMvl4pprrgEgPj6ea6+9Fr/fT0tLC2vWrAEgMzOT5uZmAJqbm8nIyCAmJoaEhAQSExPp\n7u4OZ9ciInIBWV4T6O3tpaenh+XLl9Pf34/L5QJgyZIl9Pf3AxAIBPB4POOv8Xg8+P2Rr5rzgfqd\nhrIwlIWhLKyxVAQGBwepqKjgV7/61Xjv/6yoqKhpX3u+x786cmTCz3M97u4+ND5uamqa8EbTWONI\nj9vb2201n0iO29vbIz6fw18eHh9H4vPqu+NQhbUmADAyMsKOHTtYuXIlv/jFLwAoKCigtLQUl8tF\nIBBgy5YtVFRUUF1dDUB2djYAZWVl5ObmkpKSMunv1pqAiMwnjlsTCAaDvPDCCyQlJY0XAIC0tDQa\nGhoA2Lt3L+np6ePb9+3bx8jICD6fj97eXpKTk8PZtYiIXEBhFYHPP/+cxsZGPvnkE4qLiykuLqat\nrY2cnBw6OzspKiqiq6uLnJwcAJKSksjKyqKkpITy8nLy8/PP2w6Sb333tNPplIWhLAxlYU1Y1wmk\npqby5ptvTvpYcXHxpNu9Xi9erzec3YmIyCzRFcM2p/uiGMrCUBaGsrBGRUBExMFUBGxO/U5DWRjK\nwlAW1qgIiIg4mIqAzanfaSgLQ1kYysIaFQEREQdTEbA59TsNZWEoC0NZWKMiICLiYCoCNqd+p6Es\nDGVhKAtrVARERBxMRcDm1O80lIWhLAxlYY2KgIiIg6kI2Jz6nYayMJSFoSysUREQEXEwFQGbU7/T\nUBaGsjCUhTUqAiIiDqYiYHPqdxrKwlAWhrKwRkVARMTBVARsTv1OQ1kYysJQFtaoCIiIOJiKgM2p\n32koC0NZGMrCGhUBEREHm9Mi0NHRQUlJCUVFRbz77rtzuet5S/1OQ1kYysJQFtbMWREYGxvj+eef\np7CwkO3bt1NXV0dPT89c7V5ERCYxZ0Wgu7ubxMREEhISiImJISMjg5aWlrna/bylfqehLAxlYSgL\na+asCPj9fjwez/jY7Xbj9/vnavciIjKJmEhPYCpLr1wc0f1fEhMd0f2f1dTUpH/p/JeyMJSFoSys\nmbMi4Ha76evrGx/39fXhdrunfP6Cof+Zi2lN6d/d/8O/IzqDby1evJjW1tZIT8MWlIWhLAy7ZPG9\nyyM9g/DMWRG49tpr6e3txefz4Xa7+fDDD3n00Ucnfe66devmaloiIo4WFQwGg3O1s46ODl577TVG\nR0dZt24dXq93rnYtIiKTmNMiICIi9qIrhkVEHExFQETEwSL6FdGOjg5ef/318TWC9evXn/OcN954\ng9bWVi699FLy8/P5wQ9+EIGZzr7zZdHY2Mg777wDQFJSEnfddRdXX311JKY6q2bynoBvLz588skn\neeyxx7j11lvneJZzYyZZdHd38/rrrzM4OEhcXBylpaVzP9E5cL4shoeHefHFF/nqq69YtGgRd9xx\nB+np6RGa7eyqqqri4MGDxMfHs3PnzkmfE9LnZjBCRkdHg4888kjw66+/Dp45cyZYVFQUPHLkyITn\nHDhwILh169ZgMBgMdnZ2Bjdt2hSJqc66mWTx+eefB0+ePBkMBoPB+vr6izKLmeRw9nmlpaXBbdu2\nBT/66KMIzHT2zSSLgYGB4GOPPRY8duxYMBgMBvv7+yMx1Vk3kyz+8Y9/BF966aVgMBgM+ny+4COP\nPBIcGxuLxHRnXUdHR/CLL74IPv7445M+HurnZsTaQTO5jURLSwtr1qwBICUlhZMnT3L8+PFITHdW\nzSSL5cuXs3jxtxfQ3XzzzROuubhYzPTWIu+++y633XYb8fHxEZjl3JhJFk1NTdx6663jV+JfrHnM\nJIvFixdz+vRpRkZGGBgYYOHChURFRUVoxrNrxYoVxMXFTfl4qJ+bESsCM7mNxP99jsfjuShvNRHq\nLTXef/990tLS5mJqc2qm74mWlhZ+9rOfAVy0f9FnksXRo0cZGBhg8+bNFBcX09jYONfTnBMzyWL1\n6tWMjY3xm9/8hs2bN/Pb3/52rqdpG6F+btp+YTiob7BO8Mknn9DY2MiGDRsiPZWIeO2117jnnnuI\niooiGAw6+v0xOjpKR0cHjz/+OIWFhfzlL39heHg40tOKiL///e9ER0fz4osvsnnzZrZv387Y2Fik\npxUxofy9iNjC8ExuIxHqrSbmq5ke55dffsmLL77Ipk2bpj0dnK9mksMXX3xBRUUFAN988w1tbW3E\nxMRcdGdGM8nC4/GwcuVKXC4XAD/60Y/o6Ohg5cqVczrX2TaTLDo6Oli7di2XXnopKSkpXHHFFRw9\nevSi/SLJdEK+Rc9cTGoy372NxMjICB9++OE5f5HT0tL44IMPAOjs7CQuLm78DX8xmUkWx44dY+fO\nnWzcuJHExMQIzXR2zSSH5557jsrKSiorK7ntttt44IEHLroCADPLIj09nY6ODoaGhhgYGODw4cOk\npqZGaMazZyZZ3HjjjRw4cICxsTG+/vprBgYGHFkAIPTPzYheMTzZbST++c9/AvDTn/4UgD//+c+0\ntrYSGxvLQw89RFJSUqSmO6vOl8ULL7zA/v37ufLKKwGIjo5m27ZtkZzyrJjJe+KsqqoqVq1adVF/\nRfR8Wbz33nu8++67nDlzhl/+8pf8/Oc/j+SUZ835sjh16hRvvvkmn332GfHx8axfv56bb745wrOe\nHRUVFXz66aecOHECl8vF3XffzejoKBDe56ZuGyEi4mC2XxgWEZHZoyIgIuJgKgIiIg6mIiAi4mAq\nAiIiDqYiICLiYCoCIiIOpiIgIuJg/x+wJ9GNwprPRAAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x1143863d0>"
]
}
],
"prompt_number": 87
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"RSV count by oxygen status"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"_ = hospitalized.groupby('oxygen').boxplot(column='rsv_count')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEGCAYAAACAd+UpAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIQ8AFvZ/gtDFXIXbwhQAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x11434fb50>"
]
}
],
"prompt_number": 88
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only = hospitalized[hospitalized.pcr_result___1==1]\n",
"rsv_only['vitd_lt_20'] = rsv_only.hospitalized_vitamin_d<20\n",
"rsv_only.groupby('vitd_lt_20')['oxygen'].describe()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 89,
"text": [
"vitd_lt_20 \n",
"False count 663.000000\n",
" mean 0.380090\n",
" std 0.485775\n",
" min 0.000000\n",
" 25% 0.000000\n",
" 50% 0.000000\n",
" 75% 1.000000\n",
" max 1.000000\n",
"True count 719.000000\n",
" mean 0.436718\n",
" std 0.496324\n",
" min 0.000000\n",
" 25% 0.000000\n",
" 50% 0.000000\n",
" 75% 1.000000\n",
" max 1.000000\n",
"dtype: float64"
]
}
],
"prompt_number": 89
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion on oxygen and vitamin D, by month"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_month = hospitalized.groupby('admission_month')\n",
"ax = groupby_month['oxygen'].mean().plot(grid=False, color='c', figsize=(14,8))\n",
"groupby_month['hospitalized_vitamin_d'].mean().plot(secondary_y=True, color='r')\n",
"ax.set_ylabel('Proportion on oxygen', color='c')\n",
"ax.right_ax.set_ylabel('Mean vitamin D', color='r')\n",
"ax.set_xlabel('Admission month')\n",
"ax.set_xticklabels(['Jan','Feb','Apr','Jun','Aug','Oct','Dec'])\n",
"#ax.set_xticklabels(['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec'])\n",
"_ = plt.xlim(1,12)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAA2sAAAHzCAYAAAC+H4WXAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
"<matplotlib.figure.Figure at 0x11443f950>"
]
}
],
"prompt_number": 90
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"py = plotly.plotly('jordan_ari', 'ixur4zqdut')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 91
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"c = ['#1f77b4', # muted blue\n",
" '#ff7f0e', # safety orange\n",
" '#2ca02c' # cooked asparagus green\n",
" ]\n",
"\n",
"months = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec']\n",
"\n",
"groupby_month = hospitalized.groupby('admission_month')\n",
"\n",
"data = [{'y': list(groupby_month['oxygen'].mean()), \n",
" 'x': months,\n",
" 'name': 'oxygen'},\n",
" {'y': list(groupby_month['hospitalized_vitamin_d'].mean()), \n",
" 'x': months,\n",
" 'name': 'vitamin D',\n",
" 'yaxis': 'y2'},\n",
" {'y': list(groupby_month['pcr_result___1'].mean()), \n",
" 'x': months,\n",
" 'name': 'percent RSV',\n",
" 'yaxis': 'y3'},\n",
"# {'y': list(groupby_month['pcr_result___5'].mean()), \n",
"# 'x': months,\n",
"# 'name': 'percent rhino',\n",
"# 'yaxis': 'y4'},\n",
" \n",
" ]\n",
"\n",
"layout = {\n",
" 'showlegend': False,\n",
" \"xaxis\":{\n",
" \"domain\":[0.2, 0.9]\n",
" },\n",
"\n",
" \"yaxis\":{\n",
" \"title\": 'Proportion on oxygen',\n",
" \"position\": 0.1,\n",
"\"range\":[0,1],\n",
" \"titlefont\":{\n",
" \"color\":c[0]\n",
" },\n",
" \"tickfont\":{\n",
" \"color\":c[0]\n",
" },\n",
" },\n",
"\n",
" \"yaxis2\":{ \n",
" \"overlaying\":\"y\",\n",
" \"side\":\"right\",\n",
" \"anchor\":\"x\",\n",
" \n",
" \"title\": 'Mean vitamin D',\n",
" \"titlefont\":{\n",
" \"color\":c[1]\n",
" },\n",
" \"tickfont\":{\n",
" \"color\":c[1]\n",
" }, \n",
" },\n",
"\n",
" \"yaxis3\":{ \n",
" \"overlaying\":\"y\",\n",
" \"side\":\"left\",\n",
" \"anchor\":\"free\",\n",
" \"position\": -0.15,\n",
"\"range\":[0,1],\n",
"\n",
" \"title\": \"Proportion RSV\",\n",
" \"titlefont\":{\n",
" \"color\":c[2]\n",
" },\n",
" \"tickfont\":{\n",
" \"color\":c[2]\n",
" }, \n",
" },\n",
"\n",
"# \"yaxis4\":{ \n",
"# \"overlaying\":\"y\",\n",
"# \"side\":\"left\",\n",
"# \"anchor\":\"free\",\n",
"# \"position\": -0.3,\n",
"# \"range\":[0,1],\n",
"\n",
"# \"title\": \"Proportion rhino\",\n",
"# \"titlefont\":{\n",
"# \"color\":\"#969696\"\n",
"# },\n",
"# \"tickfont\":{\n",
"# \"color\":\"#969696\"\n",
"# }, \n",
"# }\n",
"\n",
"}\n",
"\n",
"py.iplot(data, layout=layout, width=750,height=500)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"output_type": "stream",
"stream": "stdout",
"text": [
"\n",
"\n",
"\n"
]
},
{
"html": [
"<iframe height=\"550\" id=\"igraph\" scrolling=\"no\" seamless=\"seamless\" src=\"https://plot.ly/~jordan_ari/62/750/500\" width=\"800\"></iframe>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 92,
"text": [
"<IPython.core.display.HTML at 0x10bc5f8d0>"
]
}
],
"prompt_number": 92
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Risk factor tables"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['male'] = (hospitalized.sex=='M').astype(int)\n",
"age_groups = pd.get_dummies(pd.cut(hospitalized.age_months, [0,1,11,23]))\n",
"age_groups.index = hospitalized.index\n",
"age_groups.columns = 'under 2 months', '2-11 months', '12-23 months'\n",
"hospitalized = hospitalized.join(age_groups)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 93
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"nationality_lookup = {1: 'Jordanian', 2: 'Egyptian', 3: 'Palestinian', 4: 'Iraqi', 5: 'Syrian', \n",
" 6: 'Sudanese', 7: 'Russian', 8: 'Asian', 9: 'Other'}\n",
"hospitalized['nationality'] = hospitalized.mother_nationality.replace(nationality_lookup)\n",
"hospitalized['Jordanian'] = (hospitalized.nationality=='Jordanian').astype(int)\n",
"hospitalized['Palestinian'] = (hospitalized.nationality=='Palestinian').astype(int)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 94
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['vitamin D < 20'] = (hospitalized.hospitalized_vitamin_d < 20).astype(int)\n",
"hospitalized['vitamin D < 20'][hospitalized.hospitalized_vitamin_d.isnull()] = np.nan\n",
"hospitalized['vitamin D < 11'] = (hospitalized.hospitalized_vitamin_d < 11).astype(int)\n",
"hospitalized['vitamin D < 11'][hospitalized.hospitalized_vitamin_d.isnull()] = np.nan"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 95
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['premature'] = (hospitalized.gest_age < 37).astype(int)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 96
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['breastfed'] = (hospitalized.breastfed=='Breastfed').astype(int)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 97
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### ICU status"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"#hospitalized.replace({'icu_any': {True: 'ICU', False: 'No ICU'}, \n",
"# 'oxygen': {0: 'No oxygen', 1: 'Oxygen'}})['icu_any']"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 98
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_icu = hospitalized.groupby('icu_any')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 99
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def make_table(groupby, table_vars, replace_dict={}):\n",
" table = np.round(groupby[table_vars].mean(), 2).T\n",
" ratios = [calc_or(groupby, v) for v in table.index]\n",
" table['OR'] = [r[0] for r in ratios]\n",
" table['Interval'] = [r[1] for r in ratios]\n",
" table['N'] = [r[2] for r in ratios]\n",
" table.rename(columns=replace_dict, inplace=True)\n",
" table.columns.name = None\n",
" \n",
" return(table)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 100
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"table_vars = ['male', 'under 2 months', '2-11 months', '12-23 months', \n",
" 'Jordanian', 'Palestinian', 'vitamin D < 20', 'vitamin D < 11', \n",
" 'prev_cond', 'heart_hx', 'breastfed', 'premature', \n",
" 'adm_pneumo', 'adm_bronchopneumo', 'adm_sepsis', 'adm_bronchiolitis']"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 101
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Function for calculating odds ratios and simulation-based intervals."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"se = lambda p, n: np.sqrt(p * (1. - p) / n)\n",
"\n",
"def odds_ratio(x, y, n_sim=10000, alpha=0.05):\n",
"\n",
" try:\n",
" n_x, n_y = len(x.dropna()), len(y.dropna())\n",
" p_x, p_y = x.mean(), y.mean()\n",
"\n",
" se_x = se(p_x, n_x)\n",
" se_y = se(p_y, n_y)\n",
"\n",
" p_x_sim = np.random.normal(p_x, se_x, n_sim)\n",
" p_y_sim = np.random.normal(p_y, se_y, n_sim)\n",
"\n",
" ratio = ((p_x_sim / (1. - p_x_sim)) /\n",
" (p_y_sim / (1. - p_y_sim)))\n",
"\n",
" interval = np.percentile(ratio, [100*(alpha/2.), 100*(1. - alpha/2.)])\n",
"\n",
" return np.round(np.median(ratio), 2), np.round(interval, 2).tolist(), (n_y, n_x)\n",
" \n",
" except ValueError:\n",
" \n",
" return np.nan, np.nan, np.nan"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 102
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def calc_or(groupby, var):\n",
" data = list(groupby[var])\n",
" return odds_ratio(data[1][1], data[0][1])"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 103
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"virus_vars = ['RSV', 'Influenza', 'HMPV', 'Rhino']"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 104
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"pcr_lookup = {'pcr_result___1': 'RSV',\n",
"'pcr_result___2': 'HMPV',\n",
"'pcr_result___3': 'flu A',\n",
"'pcr_result___4': 'flu B',\n",
"'pcr_result___5': 'rhino',\n",
"'pcr_result___6': 'PIV1',\n",
"'pcr_result___7': 'PIV2',\n",
"'pcr_result___8': 'PIV3',\n",
"'pcr_result___13': 'H1N1',\n",
"'pcr_result___14': 'H3N2',\n",
"'pcr_result___15': 'Swine',\n",
"'pcr_result___16': 'Swine H1',\n",
"'pcr_result___17': 'flu C',\n",
"'pcr_result___18': 'Adeno'}"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 105
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['RSV'] = hospitalized['pcr_result___1']\n",
"hospitalized['Influenza'] = (hospitalized['pcr_result___3'] | hospitalized['pcr_result___4']).astype(int)\n",
"hospitalized['HMPV'] = hospitalized['pcr_result___2']\n",
"hospitalized['Rhino'] = hospitalized['pcr_result___5']"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 106
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['vitamin_d_norm'] = ((hospitalized.hospitalized_vitamin_d - hospitalized.hospitalized_vitamin_d.mean()) \n",
" / hospitalized.hospitalized_vitamin_d.std())"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 107
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized['age_centered'] = hospitalized.age_months/hospitalized.age_months.mean()"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 108
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Severity score model for bronchiolitis"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchiolitis_subset = hospitalized[hospitalized.adm_bronchiolitis==1]\n",
"bronchiolitis_subset.shape"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 109,
"text": [
"(547, 435)"
]
}
],
"prompt_number": 109
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchiolitis_subset.severity_score.hist()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 110,
"text": [
"<matplotlib.axes._subplots.AxesSubplot at 0x1140ed6d0>"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAECCAYAAAAb5qc/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBwAXI+QBwMUIeQBwMUIeAFzs/wFF4D5vaqmewwAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x112379d90>"
]
}
],
"prompt_number": 110
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchiolitis_subset = bronchiolitis_subset.dropna(subset=['vitamin_d_norm', 'male'] + virus_vars)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 111
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"import theano.tensor as T\n",
"from pymc import Bound, Normal, Uniform\n",
"\n",
"BoundedNormal = Bound(Normal, 0, 15)\n",
"\n",
"with Model() as bronchiolitis_severity_model:\n",
" \n",
" beta = Normal('beta', 0, 0.1, shape=6)\n",
" mu = BoundedNormal('mu', 4, 0.1)\n",
" \n",
" theta = mu + T.dot(bronchiolitis_subset[['vitamin_d_norm', 'male'] + virus_vars], beta)\n",
" sigma = Uniform('sigma', 0, 4)\n",
" tau = sigma ** -2\n",
" \n",
" score = BoundedNormal('score', theta, tau, observed=bronchiolitis_subset.severity_score)"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from pymc import psample\n",
"with bronchiolitis_severity_model:\n",
" trace_severity = psample(2000, NUTS(), progressbar=False) "
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"_ = forestplot(trace_severity, vars=['beta'])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"posterior_samples = {v: trace_severity.traces[0][v] for v in ['Intercept', 'vitamin_d_norm', 'male'] + virus_vars}"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchiolitis_subset.index = range(bronchiolitis_subset.shape[0])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def posterior_check(i, dataset=bronchiolitis_subset, samples=trace_severity.traces[0]):\n",
" \n",
" row = dataset.ix[i]\n",
" \n",
" pred = samples['mu']\n",
" pred += row[['vitamin_d_norm', 'male'] + virus_vars].values.dot(samples['beta'].T)\n",
"\n",
" plt.hist(pred)\n",
" plt.axvline(row.severity_score)"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"posterior_check(357)"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from pymc import Model, glm\n",
"import theano.tensor as t\n",
"from pymc import sample, Slice, NUTS \n",
"from pymc import forestplot, traceplot, summary"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"with Model() as model:\n",
" \n",
" formula = 'case ~ premature + sex_newborn + cigarette_smokers + vitamin_d_norm '\n",
" formula += '+ breastfed + prev_cond + birth_spring + birth_summer + birth_autumn + csection'\n",
" \n",
" glm.glm(formula, data_glm,\n",
" family=glm.families.Binomial(link=glm.links.Logit))"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "heading",
"level": 3,
"metadata": {},
"source": [
"Vitamin D levels for diagnoses with and without RSV"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchiolitis_subset.replace({'RSV': {1: 'RSV', 0: 'No RSV'}}).groupby('RSV').boxplot(column='hospitalized_vitamin_d');"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEHCAYAAABLKzaMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxpAnIlIwhjwRkYIx5ImIFIwhT0SkYP8PJ1D4PlgK0qgAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x1140f2410>"
]
}
],
"prompt_number": 112
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"pneumonia_subset = hospitalized[hospitalized.adm_pneumo==1]\n",
"pneumonia_subset.shape"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 113,
"text": [
"(394, 435)"
]
}
],
"prompt_number": 113
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"pneumonia_subset.replace({'RSV': {1: 'RSV', 0: 'No RSV'}}).groupby('RSV').boxplot(column='hospitalized_vitamin_d');"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEHCAYAAABLKzaMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nyAOAhf0/RNipKp/S9mQAAAAASUVORK5CYII=\n",
"text": [
"<matplotlib.figure.Figure at 0x114a17dd0>"
]
}
],
"prompt_number": 114
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchopneumo_subset = hospitalized[hospitalized.adm_bronchopneumo==1]\n",
"bronchopneumo_subset.shape"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 115,
"text": [
"(1020, 435)"
]
}
],
"prompt_number": 115
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bronchopneumo_subset.replace({'RSV': {1: 'RSV', 0: 'No RSV'}}).groupby('RSV').boxplot(column='hospitalized_vitamin_d');"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXkAAAEHCAYAAABLKzaMAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt4U/X9B/B32lKSUnpiYYWnsMqtRqj81A5wjEqjUTcU\nn5V1sEdRqZcNQZApXeh0SofTglTHFivPHqrChhdw0cwbOKlgDTjX0nKx5VJxyq000qYp9kJv398f\njKylbW49ufTk/fqL70lyzifkw5vTb749RyWEECAiIkWKCHYBRETkPwx5IiIFY8gTESkYQ56ISMEY\n8kRECsaQJyJSMIY8EZGCMeR9lJWVhYiICKxYsaLb9pMnTyIiIgLFxcWy7D8iIgJRUVFITEzE3Llz\nUVVV1e15tbW1ePjhhzFu3Dio1WokJCRg5syZeOONNwAAP/3pT3Hdddf1eoyWlhbEx8fjySef7Fet\nRJfypH/Zu4HBkPeRSqWCWq3Gn//8Zxw/ftwvx5g5cybOnDmDr7/+GiaTCSUlJZg9e3a352RmZmLb\ntm1YtWoVDh48iO3bt+OOO+5AXV0dAGDhwoUoKSnBgQMHeuzfbDajoaEBDzzwgF/qp/Dmrn/ZuwEi\nyCdZWVnCYDCI6667TsyfP9+5/cSJE0KlUolPPvnEue348ePiZz/7mRgxYoQYMWKEmDt3rjh9+rTL\n/S9YsEDcfPPN3bbl5uYKlUolGhoahBBC2O12oVKpxKZNm/rcT2dnp7j88svFkiVLejyWnp4ubr31\nVo/eL5E33PUvezdweCbvIyEEVCoV8vPz8frrr2Pv3r29Pq+5uRlpaWloaGjAtm3b8MEHH+Dbb7/F\nzJkz0dbW5vYYANDZ2Yn9+/fjzTffxNSpUzF06FAAQGxsLIYOHYoPPvgATU1Nve5DpVLh/vvvx6uv\nvoqWlhbn9qqqKhQXF+NXv/qVL2+fyC1X/cveDaAg/yczYHU9U5kzZ47Q6/VCiJ5n8oWFhUKlUonD\nhw87X3vgwAGhUqnEX//6V5f7j4qKErGxsUKtVguVSiV+8pOfiNra2m7Pe/vtt8Xw4cNFdHS0mDJl\nili2bJn4+OOPuz3n1KlTIioqqtvxjEajGDVqlOjo6OjfXwRRLzzpX/ZuYPBMvh/Ef89U1qxZg927\nd+Pdd9/t8ZyKigqMGTMGOp3OuW3y5MkYNWoUKisrXe7/hz/8Ifbv34/i4mI8+uijKC4u7vGajIwM\nnDp1Ctu3b8ecOXNQXFwMg8GAJUuWOJ+TmJiI2267DRs2bAAAtLW1YePGjbjvvvsQEcEWIP9w17/s\n3QAJ9v8yA9WCBQvETTfd5BwvWbJETJw4UXz99dfdzuQfeeQRMWbMmB6vHz16tMjJyfF4/0IIMX/+\nfDF58mS3ZzDLli0TKpVKfPPNN85t77//vlCpVOLQoUPi73//u4iMjOz2OJGcfO1f9q78+F9hP6hU\nKuefV65cidOnT+Mvf/lLt+dcddVV+Oabb3D48GHntgMHDuDUqVO46qqrPN4/AOTm5qKyshJbtmxx\n+brk5GQAQGNjo3PbrFmzkJSUhA0bNqCwsBA//vGPkZSU5PoNEvWDL/3L3vWDYP8vM1D1dqaSl5cn\nNBpNtzP55uZmcfnllwuDwSDKyspEaWmp0Ov1Ijk5WbS1tXm1fyEuzP9fc801Qgghzp49K9LT08Wm\nTZtEeXm5qKioEBs2bBCJiYniqquuEp2dnd1eu2rVKiFJkoiMjBQWi6W/fwVEfXLXv7W1tezdAGHI\n+ygrK6vHErGWlhaRlJQkIiIiui2hPHHihMjMzBQJCQlixIgRYt68eW6XUPa2fyGE2LNnj4iIiBAf\nfvihOH/+vHjsscfEtGnTRHx8vNBoNGLcuHFi0aJFve7/4pdY/NKK/M1d//7jH/9g7waISgjXd4ba\nvn073nvvPURHR+O2226DwWBAc3MzTCYTbDYbRowYgaVLl0KtVgfqhw8iIvKQyzn5pqYmbNu2DatX\nr8YzzzyDHTt24MyZMzCbzdDpdMjPz0dycjLMZnOg6iUiIi+4DPkjR45g7NixiI2NhVqtRkpKCj7/\n/HOUlpYiPT0dAKDX61FSUhKQYomIyDsuQ37ixIk4duwYbDYb7HY7ysvLUVtbC4fDAa1WCwCQJAkO\nhyMgxRIRkXeiXD2oVquxYMECvPTSS2hqasLEiRN7/ALCpcukiIgodLgMeQCYMmUKpkyZAgB4/fXX\nMWzYMEiShPr6emi1WtjtdkiS5HIfRUVF8lRL5AWDwRCU47LfKRj66ne3Ie9wOCBJEs6ePYt///vf\nePrpp2Gz2bBr1y5kZGTgk08+wdSpU90WkJqa6n3VRD4qKysL6vHZ7xRIrvrdbcg///zzaGxshFqt\nxkMPPYSYmBhkZmbCZDIhOzvbuYSSiIhCj9uQ//3vf99jm0ajgdFo9EtB9D9WqxVpaWnBLoMoYNjz\n8uO1a4iIFIwhH8J4RkPhhj0vP4Y8EZGCMeRDmNVqDXYJRAHFnpcfQ56ISMEY8iGM85MUbtjz8mPI\nExEpGEM+hHF+ksINe15+DHkiIgVjyIcwzk9SuGHPy48hH8KsVrdXnSAicokhH8Jee+10sEsgCijO\nycuPIU9EpGCcDwgxVmuUc5rmjTd0SEpqBgCkpbUjLa09mKUR+R3n5OXHkA8xl4Z5Tk5LEKshooHO\nbcjv2LEDu3btQltbGyZOnIisrCw0NzfDZDLBZrM5bxqiVqsDUW9YOX78OICEYJdBFDC8nrz8XM7J\nf/fdd3j77bfxu9/9Dnl5eaiursa+fftgNpuh0+mQn5+P5ORkmM3mQNUbViZPrg12CUQ0wLkM+ejo\naABAU1MTWltbcf78eQwZMgSlpaVIT08HAOj1epSUlPi/0jC0aNHEYJdAFFA8i5efy+ma6OhoPPDA\nA3jooYcwaNAgzJo1C8nJyXA4HNBqtQAASZLgcDgCUiwREXnH5Zl8Q0MDCgsL8cc//hEFBQU4evQo\n9u7d2+05KpXKrwWGM64ZpnDDnpefy5D/8ssvkZycjJEjR2Lo0KGYPn06Dh06BEmSUF9fDwCw2+2Q\nJMntgbp+eFarlWOO/ToOtmC/f47Da+yKSggh+nqwqakJK1asQF5eHgYPHoznn38et956K/bv34/Y\n2FhkZGTAYrGgsbER8+fP7/MgRUVFSE1NdVkIkZzKyspgMBiCcmz2OwWaq353OScfExODzMxMrF27\nFq2trbj66quRkpKCCRMmwGQyITs727mEkoiIQo/bdfJ6vR56vb7bNo1GA6PR6K+a6L+sVq4ZpvDC\nnpcfr11DRKRgDPkQxjMaCjfsefkx5ImIFIwhH8JCYSkgUSCx5+XHkCciUjCGfAjj/CSFG/a8/Bjy\nREQKxpAPYZyfpHDDnpcfQ56ISMEY8iGM85MUbtjz8mPIExEpGEM+hHF+ksINe15+DHkiIgVjyIcw\nzk9SuGHPy48hT0SkYG6vJ3/69GmsW7fOOa6pqcEvfvEL3HDDDTCZTLDZbM4bh6jVar8WG254bW0K\nN+x5+bkN+cTERDz77LMAgM7OTjz44IOYNm0azGYzdDodjEYjLBYLzGazy1sAEhFR4Hk1XXPw4EGM\nHDkSw4cPR2lpKdLT0wFcuHtUSUmJXwoMZzyjoXDDnpefVyG/e/duzJgxAwDgcDig1WoBAJIkweFw\nyF8dERH1i8ch397ejr1792L69Ok9HlOpVLIWRResX38o2CUQBRTXycvP45AvLy/HuHHjEBcXB+DC\n2Xt9fT0AwG63Q5Ikl6/v+uFZrVaOPRgfPDgspOoZSONgC/b75zi8xq6ohBDC5TP+a926dbjmmmug\n1+sBAJs3b0ZsbCwyMjJgsVjQ2NjY5xevRUVFSE1N9eQw1MXq1Wrk5LQEu4wBqaysDAaDISjHZr9T\noLnqd7erawCgpaUFBw8exMKFC53bMjMzYTKZkJ2d7VxCSf1ntUbBar3wsTz7rMa5PS2tHWlp7cEq\ni4gGKI9CXq1W46WXXuq2TaPRwGg0+qWocNY1zI8fP46cnIQgV0QUOFYr18nLjb/xSkSkYAz5EHbn\nnYnBLoEooHgWLz+GfAjjHDwR9RdDPoSFwlJAokBiz8uPIU9EpGAM+RDG+UkKN+x5+THkiYgUjCEf\nwjg/SeGGPS8/hjwRkYIx5EMY5ycp3LDn5ceQJyJSMIZ8COP8JIUb9rz8GPIh7OL15ImIfMWQD2EO\nx7XBLoEooDgnLz+GPBGRgnl805DCwkIcP34cbW1tWLx4MUaPHg2TyQSbzea8aYharfZ3vYrHm4ZQ\nOOP15OXnUcgXFhZi0qRJWLJkCTo6OnD+/HmYzWbodDoYjUZYLBaYzeY+b/9HnuNNQ4hITm6na5qa\nmnD48GHceOONAIDIyEjExMSgtLQU6enpAAC9Xo+SkhL/VhqWxgS7AKKA4lm8/NyeydtsNsTFxaGg\noABfffUVkpOTce+998LhcECr1QIAJEmCw+Hwe7FEROQdt2fyHR0dOHbsGK677jrk5eWhvb0dn332\nWbfnqFQqvxUYzmpqaoJdAlFAcZ28/NyG/LBhwxAbG4spU6YgOjoaM2bMwL59+6DValFfXw8AsNvt\nkCTJ5X66fnhWq5XjPsZWaxQWL7Zh8WIbdu5MwurVaixebMP69YdCor6BMg62YL9/jsNr7IpKCCFc\nPgPA448/jqysLIwfPx6vvPIKxowZg+rqasTGxiIjIwMWiwWNjY19fvFaVFSE1NRUd4ehS8yeHYv3\n3vsu2GUMSGVlZTAYDEE5NvudAs1Vv3u0uuahhx5CQUEBGhoakJSUhPnz50MIAZPJhOzsbOcSSuq/\nrkso9+wZhNWrLyxL5RJKIvKFRyGfmJiIp59+usd2o9Eoe0HhrmuYl5aeRU5ObJArIgocq5Xr5OXG\n33gNYQkJTcEugYgGOIZ8CLvzzsRgl0AUUDyLlx9DnohIwRjyIey1104HuwSigAqF5a9Kw5AnIlIw\nj1bXUOB0XUL5xhs6JCU1A+ASSlKW+Ph4r19TV1fnh0qUjyEfYi4N85ycliBWQ+QffQX2LYXl+OcD\nvFmOnDhdE8JKS88GuwQiGuAY8iHMgytOEBG5xJAPYVOnfi/YJRDRAMc5+RDD2/8RkZwY8iGma5i/\n/34jcnIig1wREQ1knK4hIlIwnsmHmK7TNRUVGqxezXXyROQ7hnwI6PsXQ1bi2Wd/DwB49tnuj/AX\nQ4jIEx7fNESj0SAiIgKRkZHIy8tDc3MzTCYTbDab86YharXa3/UqUl+BnTKvHhVblwW4GiJSEo/P\n5HNzcxEb+78bWJjNZuh0OhiNRlgsFpjN5j5v/0e+GTreDkAb7DKIaADz+IvXS38xp7S0FOnp6QAA\nvV6PkpISeSsjxI2vD3YJRDTAeXQmr1KpsGrVKqhUKtxyyy246aab4HA4oNVeOMuUJAkOh8OvhRIR\nkfc8CvmnnnoKl112GU6ePIm8vDyMGjWq2+MqlcovxRERUf94NF1z2WWXAQBGjx6NadOm4csvv4Qk\nSaivvzCdYLfbIUmSy310vRmA1WrlmGO/joMt2O+f4/Aau6ISbq6Cdf78eXR2dkKj0aChoQFPPvkk\n7r33Xhw8eBCxsbHIyMiAxWJBY2Njn1+8FhUVITU11WUh1BMvu+q7srIyGAyGoByb/e479rxvXPW7\n2+kah8OBtWvXAgCGDh2K2267DVdffTWuuOIKmEwmZGdnO5dQEhFRaHEb8gkJCc6Q70qj0cBoNPql\nKCIikgevXUNEpGAMeSIiBWPIExEpGEOeiEjBGPJERArGkCciUjCGPBGRgjHkiYgUjCFPRKRgDHki\nIgVjyBMRKRhDnohIwRjyREQKxpAnIlIwhjwRkYJ5dI/Xzs5O5OTkID4+Hjk5OWhubobJZILNZnPe\nMEStVvu7ViIi8pJHZ/IffPABRo8e7bxht9lshk6nQ35+PpKTk2E2m/1aJBER+cZtyNfW1qK8vBw3\n3ngjLt4OtrS0FOnp6QAAvV6PkpIS/1ZJREQ+cRvymzZtwl133YWIiP891eFwQKvVAgAkSYLD4fBf\nhURE5DOXIb93717ExcVh7NixzrP4S12cwiEiotDj8ovXI0eOYO/evSgvL0dbW5vzC1dJklBfXw+t\nVgu73Q5JktweyGq1Ii0tzflnABy7GQNDQqqegTSOiYlBMLHfOQ6VfleJvk7RL1FZWYl33nkHOTk5\n2Lx5M2JjY5GRkQGLxYLGxkbMnz+/z9cWFRUhNTXVk8NQF7cUluOfD1wb7DIGpLKyMhgMhqAcm/3u\nO/a8b1z1u1fr5C9OzWRmZuLo0aPIzs5GVVUVMjMz+18lERHJzqN18gAwadIkTJo0CQCg0WhgNBr9\nVhQREcmDv/FKRKRgDHkiIgVjyBMRKRhDnohIwRjyREQKxpAnIlIwhjwRkYIx5ImIFIwhT0SkYAx5\nIiIFY8gTESkYQ56ISMEY8kRECsaQJyJSMIY8EZGCubyefGtrK3Jzc9HW1obo6GhMnz4ds2fPdt4G\n0GazYcSIEVi6dCnUanWgaiYiIg+5DPno6GisXLkSgwcPRltbG3JycvCDH/wARUVF0Ol0MBqNsFgs\nMJvNLm//R0REweF2umbw4MEAgJaWFnR2dmLQoEEoLS1Feno6AECv16OkpMS/VRIRkU/c3v6vs7MT\nK1aswIkTJ5CVlYXhw4fD4XBAq9UCACRJgsPh8HuhRETkPbchHxERgbVr18JmsyEvLw86na7b4xdv\n7k1ERKHH49U1CQkJuPbaa1FZWQlJklBfXw8AsNvtkCTJ7eutVmu3P3PMsT/HwRbs989xeI1dUQkh\nRF8PNjQ0IDIyEkOGDMG5c+ewcuVK3Hvvvdi/fz9iY2ORkZEBi8WCxsZGl1+8FhUVITU11WUh1NMt\nheX45wPXBruMAamsrAwGgyEox2a/+4497xtX/e5yuqa+vh4FBQXo7OyEVqvF7NmzMXnyZEyYMAEm\nkwnZ2dnOJZRERBR6XIZ8UlIS1qxZ02O7RqOB0Wj0W1FKlvm3Azh3vsPj599SWO7xc4cOjoT57v/z\npSwiUii3X7ySvM6d7/D4x1Gr1Yq0tDSP9+3NfwhEFB4Y8iHMm4AnCkXe/uQK8KdXuTHkichvvPnJ\n1Rf86dU9XqAshIXCUkCiQGLPy48hT0SkYAz5EMY5eQo37Hn5MeSJiBSMIR/COD9J4YY9Lz+GPBGR\ngjHkQxjnJyncsOflx5AnIlIwhnwI4/wkhRv2vPwY8kRECsaQD2Gcn6Rww56XH0OeiEjB3F6g7OzZ\nsygoKIDD4UBcXBz0ej30ej2am5thMplgs9mcNw5Rq9WBqDlseHupYaKBjj0vP7chHxUVhQULFmDM\nmDFoaGjA8uXLMWHCBOzatQs6nQ5GoxEWiwVms9nlLQCJiCjw3Ia8VquFVqsFAMTFxWH8+PGoq6tD\naWkpcnNzAQB6vR65ubkMeQ88+ruHsP13nj9/uzf7BoAH9nhZEVHo4Fm8/Ly6nvyZM2dw8uRJXHHF\nFXA4HM7wlyQJDofDLwUqzfN/KPDb9bVvKSzHT/yyZyIaqDz+4rWlpQXr1q3DggULesy9q1Qq2Qsj\nrhmm8MOel59HZ/Lt7e147rnncP3112Pq1KkALpy919fXQ6vVwm63Q5Ikl/vo+oXKxQ+SY9fjrn93\nnj1/SEjVH8xxTEwMgon9fmHs7fQk4P0UpfXKZwP2fkJ17KrfVUII0eejAIQQKCgowNChQ7FgwQLn\n9s2bNyM2NhYZGRmwWCxobGzsc06+qKgIqamprg4TNm4pLPfrdI0/b7U2kJSVlcFgMATl2Oz3//F3\nT7LnL3DV727P5I8cOYJPP/0USUlJMBqNAIA777wTmZmZMJlMyM7Odi6hJCKi0OI25K+88kps2bKl\n18cuhj75B9cMU7hhz8uPv/FKRKRgDPkQxjMaCjfsefkx5ImIFIwhH8K4ZpjCDXtefgx5IiIFY8iH\nMM5PUrhhz8uPIU9EpGAM+RDG+UkKN+x5+THkiYgUjCEfwjg/SeGGPS8/hjwRkYIx5EMY5ycp3LDn\n5ceQJyJSMIZ8COP8JIUb9rz8vLrHK8njlsJyv+x36OBIv+yXiAYutyH/4osvory8HHFxcXjuuecA\nAM3NzTCZTLDZbM4bhlx631fqnTd3seFdbyjc8Hry8nM7XXPDDTfgscce67bNbDZDp9MhPz8fycnJ\nMJvNfiuQiIh85zbkJ06ciCFDhnTbVlpaivT0dACAXq9HSUmJf6ojorDCs3j5+fTFq8PhgFarBQBI\nkgSHwyFrUUREJI9+r65RqVRy1EFExHXyfuBTyEuShPr6egCA3W6HJEluX9P1w7NarRxz7NdxsAX7\n/XMcXmNXVEII4fIZAGw2G9asWeNcXbN582bExsYiIyMDFosFjY2NmD9/fp+vLyoqQmpqqrvD0CW4\nusZ3ZWVlMBgMQTk2+/1//N3D/Ddygat+d3smv27dOjzxxBOorq7GokWLsHPnTmRmZuLo0aPIzs5G\nVVUVMjMzZS+aiIj6z+06+V//+te9bjcajbIXQ0ThzWrlOnm58bIGREQKxpAnopDBs3j5MeSJiBSM\nFygjIr/y1wX5AF6UzxMMeSLyG2+XN3JJpPw4XUNEpGAMeSIiBWPIExEpGEOeiEjBGPJERArGkCci\nUjCGPBGRgjHkiYgUjCFPRKRgDHkiIgXr12UNKisrsWnTJnR0dMBgMGDWrFly1UVERDLw+Uy+s7MT\n69evx/Lly7F69Wp8/PHHOHnypJy1ERFRP/kc8l9++SVGjhyJhIQEREVFYcaMGSgtLZWzNiIi6ief\nQ76urg7Dhg1zjuPj41FXVydLUUREJA9+8UpEpGA+f/EaHx+P2tpa57i2thbx8fF9Pr+srMzXQ4Wt\n1an8exuo+Ln5hj0vP59Dfvz48Thz5gxsNhvi4+OxZ88eLFu2rNfnGgwGnwskGmjY7xRKVEII4euL\nKysrsXHjRucSyltvvVXO2oiIqJ/6FfJERBTa+MUrEZGCMeSJiBSMIU9EpGADOuRtNhuWL1/ul33X\n1dXh+eefBwB8/fXXKC8vd/uaiooKrF69GgBQWloKi8XS7zpyc3Px1Vdfef26J554AgDw7bffwmq1\n9ruOiz766CMUFxf3ez++vq9wx57vG3u+dwM65P0pPj4ejz76KADPG76rKVOmICMjo991qFQqn173\n1FNPAbgQCnI2/M0334yZM2f2ez++vi/yH/Z87wZ6z/frKpShQAiBwsJCVFRUYOLEibjvvvtw8uRJ\nrF+/Ht999x3GjBmDxYsXY8iQIbBardi2bRtaW1sxevRoLFu2DFu3boXdbseJEydQW1uLuXPn4sYb\nb4TNZsOaNWuwZs0abNmyBW1tbTh8+DAyMjKQkJCAjRs3orW1FUlJSfj5z3+OkSNHdqtr165d+Oqr\nr3DffffhN7/5jfMDPn36NB5//HGMHTsWL7/8MqqqqiCEwD333IPU1FS0trbixRdfRFVVFZKTk9He\n3t7ne//oo49QU1ODu+66q8cx7777bvztb3/Da6+9hlOnTsFoNEKv12Pq1Kl44YUX0NLSgoSEBMyZ\nMwfjxo1DRUUFzGYzYmJi8M0338BgMCAxMRFvvfUW4uLisHDhQgwbNgxbt26FRqPB7bffjtzcXKSk\npGDv3r0YPHgwsrKyMHbs2F5r9eZ9kWvsefa8V8QAVlNTI+bNmycOHDggOjo6xB/+8AdRUVEh1q5d\nK3bv3i3a29vFhg0bxCuvvCKEEGLZsmWipaVFCCFEY2OjEEKILVu2iIcffljU19eL6upqsWjRItHZ\n2SlqamrEo48+KoQQYufOneKll15yHrepqUl0dHQIIYTYvXu3yM/PF0II8cUXX4i8vLxeXyOEECUl\nJeLJJ58U7e3t4tVXXxXvv/++EEIIu90ufvvb3wohhPjXv/4l8vPzRWtrqzhw4ICYN2+eOHbsWK/v\n3+FwiKVLlzrHTz/9tDh8+LAQQoi7775bCCFERUWFsyYhhDh//rxobW0VQghx9OhRsWLFCmftd9xx\nh6iurhbNzc0iKytLvPzyy6Kjo0Ns3bpVvPPOO0IIIbZu3SreffddIYQQubm5oqCgQHR0dIji4mJR\nUFDQ52flzfuivrHn2fPeGvBn8vHx8Zg8eTIAYNKkSaisrMSxY8ewfPlyqFQq6PV6bNiwAQAwbtw4\n/OlPf8LMmTMxbdo0ABd+hLrmmmsgSRIkSUJSUhKOHDni8hINra2teP3113Ho0CEIIdDR0eG2zurq\narz66qtYuXIlIiMjceDAAbS1tWHXrl0AgMbGRtTU1GDfvn1IS0vDoEGDMHnyZAwfPrzPfcbFxWHE\niBGoqqrCyJEjcfr0aeh0um7PEb38GsSWLVvwxRdfoLOzE9XV1c7tEyZMcJ6djR49GlOnTkVERAR0\nOh0++uijXmu4/vrrERERgZSUFLz11lt91urN+yLX2PPseW8M+JCPiYlx/jkqKgotLS0ALnzQl86B\nPfzwwzhy5Ag+/fRTvPPOO3jmmWcghOjRFO7mzj788EMMHToUeXl5OHnyJNauXevy+S0tLVi3bh0W\nLlwIrVbr3H7//fdj0qRJPZ7fW5P25Uc/+hE+++wzJCYmOv8Ru7Jnzx6cO3cOq1atwvnz5/HLX/7S\n+dilf5cXx5GRkWhra+t1f0OGDHE+v7W11eWxvXlf1Df2PHveG4r74jUqKgrjx4/H559/jo6ODuza\ntQspKSkQQsBms0Gn0+Gee+6B3W53fkD79+9HQ0MDampqcOLECSQnJ3fbp0ajQUNDg3NcV1eHhIQE\nAMCOHTvc1vTiiy9Cr9fjyiuvdG67+uqrsWPHDjQ3NwMA/vOf/wAArr32WuzZswdtbW344osvcPbs\nWZf7njZtGkpKSrB7927MmDGjx+MajQbnzp3rVvv3vvc9DBo0CEVFRV43YW8B4Qlv3xd5jj3fHXu+\nuwF/Jn/pGYhKpcLcuXOxfv16bN68GWPHjsXixYvR0dGBF154AU1NTdBoNJg7dy6io6OhUqmQkpKC\nNWvWoLa2FvPmzUNERES3faekpODjjz+G0WjEnDlzMGvWLBQWFuK9997DjBkzutVwaT1nz57F559/\njjNnzmAOP561AAAA7klEQVTnzp0AgAcffBCZmZnYuHEjsrOzoVarkZCQgBUrVjgb45FHHsGECRNw\nxRVXuHz/Q4YMwahRo3Dq1CmMHz++Rx2XX345hg8f7vwSKj09HevXr0d2djamT58OtVrdZ+1dt198\nrOuf3X0WXXn7vqhv7Hn2vDfC/to1b775JtRqNW6//fZgl0IUEOz58KK46RpfcM02hRv2fPgI+zP5\ngeLxxx/v8UXQ0qVL8f3vfz9IFfVu3759eO2117ptS0hIQHZ2dpAqooGKPS8PhjwRkYJxuoaISMEY\n8kRECsaQJyJSMIY8EZGCMeSJiBTs/wHdeiqQZFy+SQAAAABJRU5ErkJggg==\n",
"text": [
"<matplotlib.figure.Figure at 0x114456b50>"
]
}
],
"prompt_number": 116
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Oxygen"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_o2 = hospitalized.groupby('oxygen')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 96
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(groupby_o2, table_vars=table_vars+virus_vars, replace_dict={0.0: 'No Oxygen', 1.0: 'Oxygen'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>No Oxygen</th>\n",
" <th>Oxygen</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.62</td>\n",
" <td> 0.58</td>\n",
" <td> 0.86</td>\n",
" <td> [0.74, 1.01]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.17</td>\n",
" <td> 0.23</td>\n",
" <td> 1.47</td>\n",
" <td> [1.23, 1.77]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.54</td>\n",
" <td> 0.53</td>\n",
" <td> 0.96</td>\n",
" <td> [0.82, 1.11]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.18</td>\n",
" <td> 0.11</td>\n",
" <td> 0.57</td>\n",
" <td> [0.45, 0.71]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.89</td>\n",
" <td> 0.91</td>\n",
" <td> 1.20</td>\n",
" <td> [0.95, 1.58]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.08</td>\n",
" <td> 0.06</td>\n",
" <td> 0.75</td>\n",
" <td> [0.54, 1.01]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.57</td>\n",
" <td> 0.60</td>\n",
" <td> 1.13</td>\n",
" <td> [0.96, 1.34]</td>\n",
" <td> (1807, 858)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.36</td>\n",
" <td> 0.43</td>\n",
" <td> 1.30</td>\n",
" <td> [1.1, 1.54]</td>\n",
" <td> (1807, 858)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.09</td>\n",
" <td> 0.18</td>\n",
" <td> 2.16</td>\n",
" <td> [1.72, 2.68]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.03</td>\n",
" <td> 0.07</td>\n",
" <td> 2.06</td>\n",
" <td> [1.46, 2.88]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.62</td>\n",
" <td> 0.59</td>\n",
" <td> 0.88</td>\n",
" <td> [0.76, 1.03]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.13</td>\n",
" <td> 0.17</td>\n",
" <td> 1.38</td>\n",
" <td> [1.12, 1.7]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.09</td>\n",
" <td> 0.21</td>\n",
" <td> 2.77</td>\n",
" <td> [2.25, 3.45]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.34</td>\n",
" <td> 0.28</td>\n",
" <td> 0.73</td>\n",
" <td> [0.62, 0.86]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.33</td>\n",
" <td> 0.20</td>\n",
" <td> 0.51</td>\n",
" <td> [0.42, 0.61]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.15</td>\n",
" <td> 0.21</td>\n",
" <td> 1.49</td>\n",
" <td> [1.22, 1.8]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>RSV</th>\n",
" <td> 0.38</td>\n",
" <td> 0.56</td>\n",
" <td> 2.03</td>\n",
" <td> [1.75, 2.36]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Influenza</th>\n",
" <td> 0.04</td>\n",
" <td> 0.03</td>\n",
" <td> 0.73</td>\n",
" <td> [0.42, 1.11]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>HMPV</th>\n",
" <td> 0.09</td>\n",
" <td> 0.08</td>\n",
" <td> 0.89</td>\n",
" <td> [0.67, 1.16]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Rhino</th>\n",
" <td> 0.39</td>\n",
" <td> 0.38</td>\n",
" <td> 0.95</td>\n",
" <td> [0.82, 1.11]</td>\n",
" <td> (2125, 1013)</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>20 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 97,
"text": [
" No Oxygen Oxygen OR Interval N\n",
"male 0.62 0.58 0.86 [0.74, 1.01] (2125, 1013)\n",
"under 2 months 0.17 0.23 1.47 [1.23, 1.77] (2125, 1013)\n",
"2-11 months 0.54 0.53 0.96 [0.82, 1.11] (2125, 1013)\n",
"12-23 months 0.18 0.11 0.57 [0.45, 0.71] (2125, 1013)\n",
"Jordanian 0.89 0.91 1.20 [0.95, 1.58] (2125, 1013)\n",
"Palestinian 0.08 0.06 0.75 [0.54, 1.01] (2125, 1013)\n",
"vitamin D < 20 0.57 0.60 1.13 [0.96, 1.34] (1807, 858)\n",
"vitamin D < 11 0.36 0.43 1.30 [1.1, 1.54] (1807, 858)\n",
"prev_cond 0.09 0.18 2.16 [1.72, 2.68] (2125, 1013)\n",
"heart_hx 0.03 0.07 2.06 [1.46, 2.88] (2125, 1013)\n",
"breastfed 0.62 0.59 0.88 [0.76, 1.03] (2125, 1013)\n",
"premature 0.13 0.17 1.38 [1.12, 1.7] (2125, 1013)\n",
"adm_pneumo 0.09 0.21 2.77 [2.25, 3.45] (2125, 1013)\n",
"adm_bronchopneumo 0.34 0.28 0.73 [0.62, 0.86] (2125, 1013)\n",
"adm_sepsis 0.33 0.20 0.51 [0.42, 0.61] (2125, 1013)\n",
"adm_bronchiolitis 0.15 0.21 1.49 [1.22, 1.8] (2125, 1013)\n",
"RSV 0.38 0.56 2.03 [1.75, 2.36] (2125, 1013)\n",
"Influenza 0.04 0.03 0.73 [0.42, 1.11] (2125, 1013)\n",
"HMPV 0.09 0.08 0.89 [0.67, 1.16] (2125, 1013)\n",
"Rhino 0.39 0.38 0.95 [0.82, 1.11] (2125, 1013)\n",
"\n",
"[20 rows x 5 columns]"
]
}
],
"prompt_number": 97
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Mechanical ventilation"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_vent = hospitalized.groupby('vent')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 98
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(groupby_vent, table_vars=table_vars+virus_vars, \n",
" replace_dict={0.0: 'No Ventilator', 1.0: 'Ventilator'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>No Ventilator</th>\n",
" <th>Ventilator</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.60</td>\n",
" <td> 0.59</td>\n",
" <td> 0.93</td>\n",
" <td> [0.63, 1.38]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.19</td>\n",
" <td> 0.24</td>\n",
" <td> 1.36</td>\n",
" <td> [0.83, 2.06]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.54</td>\n",
" <td> 0.47</td>\n",
" <td> 0.77</td>\n",
" <td> [0.52, 1.12]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.15</td>\n",
" <td> 0.14</td>\n",
" <td> 0.93</td>\n",
" <td> [0.46, 1.48]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.89</td>\n",
" <td> 0.94</td>\n",
" <td> 1.76</td>\n",
" <td> [0.94, 6.27]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.07</td>\n",
" <td> 0.04</td>\n",
" <td> 0.48</td>\n",
" <td> [0.02, 1.01]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.58</td>\n",
" <td> 0.50</td>\n",
" <td> 0.70</td>\n",
" <td> [0.47, 1.02]</td>\n",
" <td> (2557, 107)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.39</td>\n",
" <td> 0.35</td>\n",
" <td> 0.84</td>\n",
" <td> [0.54, 1.23]</td>\n",
" <td> (2557, 107)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.12</td>\n",
" <td> 0.28</td>\n",
" <td> 2.99</td>\n",
" <td> [1.84, 4.42]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.04</td>\n",
" <td> 0.14</td>\n",
" <td> 3.51</td>\n",
" <td> [1.7, 5.84]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.61</td>\n",
" <td> 0.59</td>\n",
" <td> 0.94</td>\n",
" <td> [0.65, 1.42]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.14</td>\n",
" <td> 0.18</td>\n",
" <td> 1.35</td>\n",
" <td> [0.75, 2.11]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.12</td>\n",
" <td> 0.23</td>\n",
" <td> 2.23</td>\n",
" <td> [1.32, 3.39]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.32</td>\n",
" <td> 0.33</td>\n",
" <td> 1.06</td>\n",
" <td> [0.7, 1.56]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.29</td>\n",
" <td> 0.22</td>\n",
" <td> 0.68</td>\n",
" <td> [0.4, 1.03]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.17</td>\n",
" <td> 0.19</td>\n",
" <td> 1.12</td>\n",
" <td> [0.62, 1.73]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>RSV</th>\n",
" <td> 0.44</td>\n",
" <td> 0.47</td>\n",
" <td> 1.13</td>\n",
" <td> [0.76, 1.65]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Influenza</th>\n",
" <td> 0.03</td>\n",
" <td> 0.02</td>\n",
" <td> 0.53</td>\n",
" <td> [-0.19, 1.33]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>HMPV</th>\n",
" <td> 0.09</td>\n",
" <td> 0.06</td>\n",
" <td> 0.71</td>\n",
" <td> [0.18, 1.29]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Rhino</th>\n",
" <td> 0.39</td>\n",
" <td> 0.37</td>\n",
" <td> 0.91</td>\n",
" <td> [0.6, 1.33]</td>\n",
" <td> (3026, 111)</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>20 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 99,
"text": [
" No Ventilator Ventilator OR Interval N\n",
"male 0.60 0.59 0.93 [0.63, 1.38] (3026, 111)\n",
"under 2 months 0.19 0.24 1.36 [0.83, 2.06] (3026, 111)\n",
"2-11 months 0.54 0.47 0.77 [0.52, 1.12] (3026, 111)\n",
"12-23 months 0.15 0.14 0.93 [0.46, 1.48] (3026, 111)\n",
"Jordanian 0.89 0.94 1.76 [0.94, 6.27] (3026, 111)\n",
"Palestinian 0.07 0.04 0.48 [0.02, 1.01] (3026, 111)\n",
"vitamin D < 20 0.58 0.50 0.70 [0.47, 1.02] (2557, 107)\n",
"vitamin D < 11 0.39 0.35 0.84 [0.54, 1.23] (2557, 107)\n",
"prev_cond 0.12 0.28 2.99 [1.84, 4.42] (3026, 111)\n",
"heart_hx 0.04 0.14 3.51 [1.7, 5.84] (3026, 111)\n",
"breastfed 0.61 0.59 0.94 [0.65, 1.42] (3026, 111)\n",
"premature 0.14 0.18 1.35 [0.75, 2.11] (3026, 111)\n",
"adm_pneumo 0.12 0.23 2.23 [1.32, 3.39] (3026, 111)\n",
"adm_bronchopneumo 0.32 0.33 1.06 [0.7, 1.56] (3026, 111)\n",
"adm_sepsis 0.29 0.22 0.68 [0.4, 1.03] (3026, 111)\n",
"adm_bronchiolitis 0.17 0.19 1.12 [0.62, 1.73] (3026, 111)\n",
"RSV 0.44 0.47 1.13 [0.76, 1.65] (3026, 111)\n",
"Influenza 0.03 0.02 0.53 [-0.19, 1.33] (3026, 111)\n",
"HMPV 0.09 0.06 0.71 [0.18, 1.29] (3026, 111)\n",
"Rhino 0.39 0.37 0.91 [0.6, 1.33] (3026, 111)\n",
"\n",
"[20 rows x 5 columns]"
]
}
],
"prompt_number": 99
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### Death"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death = hospitalized.groupby('death')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 100
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(groupby_death, table_vars=table_vars+virus_vars, replace_dict={False: 'Survived', True: 'Die'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Survived</th>\n",
" <th>Die</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.60</td>\n",
" <td> 0.48</td>\n",
" <td> 0.61</td>\n",
" <td> [0.3, 1.25]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.19</td>\n",
" <td> 0.39</td>\n",
" <td> 2.68</td>\n",
" <td> [1.15, 5.39]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.53</td>\n",
" <td> 0.39</td>\n",
" <td> 0.55</td>\n",
" <td> [0.24, 1.09]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.15</td>\n",
" <td> 0.06</td>\n",
" <td> 0.38</td>\n",
" <td> [-0.13, 0.97]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.90</td>\n",
" <td> 0.87</td>\n",
" <td> 0.78</td>\n",
" <td> [0.31, 4.99]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.07</td>\n",
" <td> 0.10</td>\n",
" <td> 1.37</td>\n",
" <td> [-0.11, 3.31]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.58</td>\n",
" <td> 0.44</td>\n",
" <td> 0.56</td>\n",
" <td> [0.24, 1.26]</td>\n",
" <td> (2664, 25)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.39</td>\n",
" <td> 0.28</td>\n",
" <td> 0.62</td>\n",
" <td> [0.18, 1.34]</td>\n",
" <td> (2664, 25)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.12</td>\n",
" <td> 0.52</td>\n",
" <td> 7.99</td>\n",
" <td> [3.78, 16.93]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.05</td>\n",
" <td> 0.13</td>\n",
" <td> 3.14</td>\n",
" <td> [0.21, 7.04]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.61</td>\n",
" <td> 0.42</td>\n",
" <td> 0.47</td>\n",
" <td> [0.21, 0.94]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.14</td>\n",
" <td> 0.16</td>\n",
" <td> 1.17</td>\n",
" <td> [0.2, 2.49]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.12</td>\n",
" <td> 0.32</td>\n",
" <td> 3.44</td>\n",
" <td> [1.37, 6.92]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.32</td>\n",
" <td> 0.13</td>\n",
" <td> 0.31</td>\n",
" <td> [0.03, 0.69]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.28</td>\n",
" <td> 0.52</td>\n",
" <td> 2.71</td>\n",
" <td> [1.33, 5.83]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.17</td>\n",
" <td> 0.03</td>\n",
" <td> 0.16</td>\n",
" <td> [-0.14, 0.5]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>RSV</th>\n",
" <td> 0.44</td>\n",
" <td> 0.23</td>\n",
" <td> 0.37</td>\n",
" <td> [0.11, 0.74]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Influenza</th>\n",
" <td> 0.03</td>\n",
" <td> 0.13</td>\n",
" <td> 4.51</td>\n",
" <td> [0.27, 10.3]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>HMPV</th>\n",
" <td> 0.09</td>\n",
" <td> 0.06</td>\n",
" <td> 0.73</td>\n",
" <td> [-0.23, 1.91]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Rhino</th>\n",
" <td> 0.39</td>\n",
" <td> 0.42</td>\n",
" <td> 1.13</td>\n",
" <td> [0.5, 2.27]</td>\n",
" <td> (3138, 31)</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>20 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 101,
"text": [
" Survived Die OR Interval N\n",
"male 0.60 0.48 0.61 [0.3, 1.25] (3138, 31)\n",
"under 2 months 0.19 0.39 2.68 [1.15, 5.39] (3138, 31)\n",
"2-11 months 0.53 0.39 0.55 [0.24, 1.09] (3138, 31)\n",
"12-23 months 0.15 0.06 0.38 [-0.13, 0.97] (3138, 31)\n",
"Jordanian 0.90 0.87 0.78 [0.31, 4.99] (3138, 31)\n",
"Palestinian 0.07 0.10 1.37 [-0.11, 3.31] (3138, 31)\n",
"vitamin D < 20 0.58 0.44 0.56 [0.24, 1.26] (2664, 25)\n",
"vitamin D < 11 0.39 0.28 0.62 [0.18, 1.34] (2664, 25)\n",
"prev_cond 0.12 0.52 7.99 [3.78, 16.93] (3138, 31)\n",
"heart_hx 0.05 0.13 3.14 [0.21, 7.04] (3138, 31)\n",
"breastfed 0.61 0.42 0.47 [0.21, 0.94] (3138, 31)\n",
"premature 0.14 0.16 1.17 [0.2, 2.49] (3138, 31)\n",
"adm_pneumo 0.12 0.32 3.44 [1.37, 6.92] (3138, 31)\n",
"adm_bronchopneumo 0.32 0.13 0.31 [0.03, 0.69] (3138, 31)\n",
"adm_sepsis 0.28 0.52 2.71 [1.33, 5.83] (3138, 31)\n",
"adm_bronchiolitis 0.17 0.03 0.16 [-0.14, 0.5] (3138, 31)\n",
"RSV 0.44 0.23 0.37 [0.11, 0.74] (3138, 31)\n",
"Influenza 0.03 0.13 4.51 [0.27, 10.3] (3138, 31)\n",
"HMPV 0.09 0.06 0.73 [-0.23, 1.91] (3138, 31)\n",
"Rhino 0.39 0.42 1.13 [0.5, 2.27] (3138, 31)\n",
"\n",
"[20 rows x 5 columns]"
]
}
],
"prompt_number": 101
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### RSV-only subset\n",
"\n",
"The following calculations repeat the above tables, but for the RSV subset."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only = hospitalized[hospitalized.pcr_result___1==1]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 102
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_icu = rsv_only.groupby('icu_any')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 103
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"virus_vars"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 104,
"text": [
"['RSV', 'Influenza', 'HMPV', 'Rhino']"
]
}
],
"prompt_number": 104
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(rsv_icu, table_vars=table_vars+virus_vars[1:], replace_dict={False: 'No ICU', True: 'ICU'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>No ICU</th>\n",
" <th>ICU</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.60</td>\n",
" <td> 0.56</td>\n",
" <td> 0.86</td>\n",
" <td> [0.61, 1.23]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.19</td>\n",
" <td> 0.32</td>\n",
" <td> 2.10</td>\n",
" <td> [1.4, 3.01]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.61</td>\n",
" <td> 0.38</td>\n",
" <td> 0.39</td>\n",
" <td> [0.27, 0.56]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.12</td>\n",
" <td> 0.06</td>\n",
" <td> 0.48</td>\n",
" <td> [0.17, 0.84]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.90</td>\n",
" <td> 0.89</td>\n",
" <td> 0.88</td>\n",
" <td> [0.53, 1.77]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.07</td>\n",
" <td> 0.08</td>\n",
" <td> 1.17</td>\n",
" <td> [0.48, 2.06]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.60</td>\n",
" <td> 0.66</td>\n",
" <td> 1.28</td>\n",
" <td> [0.89, 1.95]</td>\n",
" <td> (1072, 121)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.44</td>\n",
" <td> 0.50</td>\n",
" <td> 1.24</td>\n",
" <td> [0.84, 1.82]</td>\n",
" <td> (1072, 121)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.07</td>\n",
" <td> 0.12</td>\n",
" <td> 1.87</td>\n",
" <td> [0.96, 3.08]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.03</td>\n",
" <td> 0.03</td>\n",
" <td> 0.97</td>\n",
" <td> [0.04, 2.18]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.64</td>\n",
" <td> 0.58</td>\n",
" <td> 0.77</td>\n",
" <td> [0.54, 1.11]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.12</td>\n",
" <td> 0.20</td>\n",
" <td> 1.89</td>\n",
" <td> [1.16, 2.86]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.14</td>\n",
" <td> 0.32</td>\n",
" <td> 2.88</td>\n",
" <td> [1.91, 4.19]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.36</td>\n",
" <td> 0.14</td>\n",
" <td> 0.29</td>\n",
" <td> [0.16, 0.44]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.15</td>\n",
" <td> 0.42</td>\n",
" <td> 4.04</td>\n",
" <td> [2.79, 5.8]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.28</td>\n",
" <td> 0.17</td>\n",
" <td> 0.53</td>\n",
" <td> [0.31, 0.79]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Influenza</th>\n",
" <td> 0.02</td>\n",
" <td> 0.02</td>\n",
" <td> 0.91</td>\n",
" <td> [-0.11, 2.23]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>HMPV</th>\n",
" <td> 0.05</td>\n",
" <td> 0.07</td>\n",
" <td> 1.47</td>\n",
" <td> [0.54, 2.68]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Rhino</th>\n",
" <td> 0.33</td>\n",
" <td> 0.32</td>\n",
" <td> 0.97</td>\n",
" <td> [0.65, 1.37]</td>\n",
" <td> (1255, 142)</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>19 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 105,
"text": [
" No ICU ICU OR Interval N\n",
"male 0.60 0.56 0.86 [0.61, 1.23] (1255, 142)\n",
"under 2 months 0.19 0.32 2.10 [1.4, 3.01] (1255, 142)\n",
"2-11 months 0.61 0.38 0.39 [0.27, 0.56] (1255, 142)\n",
"12-23 months 0.12 0.06 0.48 [0.17, 0.84] (1255, 142)\n",
"Jordanian 0.90 0.89 0.88 [0.53, 1.77] (1255, 142)\n",
"Palestinian 0.07 0.08 1.17 [0.48, 2.06] (1255, 142)\n",
"vitamin D < 20 0.60 0.66 1.28 [0.89, 1.95] (1072, 121)\n",
"vitamin D < 11 0.44 0.50 1.24 [0.84, 1.82] (1072, 121)\n",
"prev_cond 0.07 0.12 1.87 [0.96, 3.08] (1255, 142)\n",
"heart_hx 0.03 0.03 0.97 [0.04, 2.18] (1255, 142)\n",
"breastfed 0.64 0.58 0.77 [0.54, 1.11] (1255, 142)\n",
"premature 0.12 0.20 1.89 [1.16, 2.86] (1255, 142)\n",
"adm_pneumo 0.14 0.32 2.88 [1.91, 4.19] (1255, 142)\n",
"adm_bronchopneumo 0.36 0.14 0.29 [0.16, 0.44] (1255, 142)\n",
"adm_sepsis 0.15 0.42 4.04 [2.79, 5.8] (1255, 142)\n",
"adm_bronchiolitis 0.28 0.17 0.53 [0.31, 0.79] (1255, 142)\n",
"Influenza 0.02 0.02 0.91 [-0.11, 2.23] (1255, 142)\n",
"HMPV 0.05 0.07 1.47 [0.54, 2.68] (1255, 142)\n",
"Rhino 0.33 0.32 0.97 [0.65, 1.37] (1255, 142)\n",
"\n",
"[19 rows x 5 columns]"
]
}
],
"prompt_number": 105
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_o2 = rsv_only.groupby('oxygen')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 106
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(rsv_o2, table_vars=table_vars, replace_dict={0.0: 'No Oxygen', 1.0: 'Oxygen'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>No Oxygen</th>\n",
" <th>Oxygen</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.61</td>\n",
" <td> 0.57</td>\n",
" <td> 0.82</td>\n",
" <td> [0.66, 1.02]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.15</td>\n",
" <td> 0.28</td>\n",
" <td> 2.27</td>\n",
" <td> [1.74, 3.0]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.63</td>\n",
" <td> 0.52</td>\n",
" <td> 0.62</td>\n",
" <td> [0.5, 0.77]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.16</td>\n",
" <td> 0.06</td>\n",
" <td> 0.34</td>\n",
" <td> [0.22, 0.49]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.90</td>\n",
" <td> 0.90</td>\n",
" <td> 0.95</td>\n",
" <td> [0.67, 1.37]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.07</td>\n",
" <td> 0.07</td>\n",
" <td> 0.97</td>\n",
" <td> [0.61, 1.47]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.58</td>\n",
" <td> 0.65</td>\n",
" <td> 1.34</td>\n",
" <td> [1.06, 1.71]</td>\n",
" <td> (697, 483)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.41</td>\n",
" <td> 0.51</td>\n",
" <td> 1.49</td>\n",
" <td> [1.18, 1.89]</td>\n",
" <td> (697, 483)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.06</td>\n",
" <td> 0.08</td>\n",
" <td> 1.33</td>\n",
" <td> [0.88, 2.03]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.03</td>\n",
" <td> 0.03</td>\n",
" <td> 1.12</td>\n",
" <td> [0.54, 2.19]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.65</td>\n",
" <td> 0.62</td>\n",
" <td> 0.89</td>\n",
" <td> [0.72, 1.12]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.11</td>\n",
" <td> 0.15</td>\n",
" <td> 1.44</td>\n",
" <td> [1.04, 2.0]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.12</td>\n",
" <td> 0.23</td>\n",
" <td> 2.19</td>\n",
" <td> [1.64, 2.95]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.42</td>\n",
" <td> 0.22</td>\n",
" <td> 0.40</td>\n",
" <td> [0.31, 0.5]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.15</td>\n",
" <td> 0.23</td>\n",
" <td> 1.71</td>\n",
" <td> [1.3, 2.25]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.27</td>\n",
" <td> 0.27</td>\n",
" <td> 0.98</td>\n",
" <td> [0.77, 1.24]</td>\n",
" <td> (816, 566)</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>16 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 107,
"text": [
" No Oxygen Oxygen OR Interval N\n",
"male 0.61 0.57 0.82 [0.66, 1.02] (816, 566)\n",
"under 2 months 0.15 0.28 2.27 [1.74, 3.0] (816, 566)\n",
"2-11 months 0.63 0.52 0.62 [0.5, 0.77] (816, 566)\n",
"12-23 months 0.16 0.06 0.34 [0.22, 0.49] (816, 566)\n",
"Jordanian 0.90 0.90 0.95 [0.67, 1.37] (816, 566)\n",
"Palestinian 0.07 0.07 0.97 [0.61, 1.47] (816, 566)\n",
"vitamin D < 20 0.58 0.65 1.34 [1.06, 1.71] (697, 483)\n",
"vitamin D < 11 0.41 0.51 1.49 [1.18, 1.89] (697, 483)\n",
"prev_cond 0.06 0.08 1.33 [0.88, 2.03] (816, 566)\n",
"heart_hx 0.03 0.03 1.12 [0.54, 2.19] (816, 566)\n",
"breastfed 0.65 0.62 0.89 [0.72, 1.12] (816, 566)\n",
"premature 0.11 0.15 1.44 [1.04, 2.0] (816, 566)\n",
"adm_pneumo 0.12 0.23 2.19 [1.64, 2.95] (816, 566)\n",
"adm_bronchopneumo 0.42 0.22 0.40 [0.31, 0.5] (816, 566)\n",
"adm_sepsis 0.15 0.23 1.71 [1.3, 2.25] (816, 566)\n",
"adm_bronchiolitis 0.27 0.27 0.98 [0.77, 1.24] (816, 566)\n",
"\n",
"[16 rows x 5 columns]"
]
}
],
"prompt_number": 107
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_vent = rsv_only.groupby('vent')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 108
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(rsv_vent, table_vars=table_vars, \n",
" replace_dict={0.0: 'No Ventilator', 1.0: 'Ventilator'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>No Ventilator</th>\n",
" <th>Ventilator</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.60</td>\n",
" <td> 0.56</td>\n",
" <td> 0.86</td>\n",
" <td> [0.49, 1.55]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.20</td>\n",
" <td> 0.33</td>\n",
" <td> 1.97</td>\n",
" <td> [1.02, 3.5]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.59</td>\n",
" <td> 0.46</td>\n",
" <td> 0.60</td>\n",
" <td> [0.33, 1.05]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.12</td>\n",
" <td> 0.08</td>\n",
" <td> 0.62</td>\n",
" <td> [0.05, 1.32]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.90</td>\n",
" <td> 0.94</td>\n",
" <td> 1.76</td>\n",
" <td> [-6.83, 14.84]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.07</td>\n",
" <td> 0.04</td>\n",
" <td> 0.55</td>\n",
" <td> [-0.18, 1.4]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.62</td>\n",
" <td> 0.48</td>\n",
" <td> 0.58</td>\n",
" <td> [0.32, 1.02]</td>\n",
" <td> (1127, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.45</td>\n",
" <td> 0.38</td>\n",
" <td> 0.75</td>\n",
" <td> [0.41, 1.3]</td>\n",
" <td> (1127, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.07</td>\n",
" <td> 0.10</td>\n",
" <td> 1.40</td>\n",
" <td> [0.22, 2.95]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.03</td>\n",
" <td> 0.00</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.64</td>\n",
" <td> 0.52</td>\n",
" <td> 0.60</td>\n",
" <td> [0.34, 1.07]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.13</td>\n",
" <td> 0.13</td>\n",
" <td> 1.08</td>\n",
" <td> [0.31, 2.08]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.16</td>\n",
" <td> 0.27</td>\n",
" <td> 1.96</td>\n",
" <td> [0.92, 3.51]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.34</td>\n",
" <td> 0.25</td>\n",
" <td> 0.64</td>\n",
" <td> [0.29, 1.13]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.18</td>\n",
" <td> 0.21</td>\n",
" <td> 1.24</td>\n",
" <td> [0.53, 2.26]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.27</td>\n",
" <td> 0.27</td>\n",
" <td> 1.00</td>\n",
" <td> [0.47, 1.77]</td>\n",
" <td> (1329, 52)</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>16 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 109,
"text": [
" No Ventilator Ventilator OR Interval N\n",
"male 0.60 0.56 0.86 [0.49, 1.55] (1329, 52)\n",
"under 2 months 0.20 0.33 1.97 [1.02, 3.5] (1329, 52)\n",
"2-11 months 0.59 0.46 0.60 [0.33, 1.05] (1329, 52)\n",
"12-23 months 0.12 0.08 0.62 [0.05, 1.32] (1329, 52)\n",
"Jordanian 0.90 0.94 1.76 [-6.83, 14.84] (1329, 52)\n",
"Palestinian 0.07 0.04 0.55 [-0.18, 1.4] (1329, 52)\n",
"vitamin D < 20 0.62 0.48 0.58 [0.32, 1.02] (1127, 52)\n",
"vitamin D < 11 0.45 0.38 0.75 [0.41, 1.3] (1127, 52)\n",
"prev_cond 0.07 0.10 1.40 [0.22, 2.95] (1329, 52)\n",
"heart_hx 0.03 0.00 NaN NaN NaN\n",
"breastfed 0.64 0.52 0.60 [0.34, 1.07] (1329, 52)\n",
"premature 0.13 0.13 1.08 [0.31, 2.08] (1329, 52)\n",
"adm_pneumo 0.16 0.27 1.96 [0.92, 3.51] (1329, 52)\n",
"adm_bronchopneumo 0.34 0.25 0.64 [0.29, 1.13] (1329, 52)\n",
"adm_sepsis 0.18 0.21 1.24 [0.53, 2.26] (1329, 52)\n",
"adm_bronchiolitis 0.27 0.27 1.00 [0.47, 1.77] (1329, 52)\n",
"\n",
"[16 rows x 5 columns]"
]
}
],
"prompt_number": 109
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_death = rsv_only.groupby('death')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 110
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"make_table(rsv_death, table_vars=table_vars, replace_dict={False: 'Survived', True: 'Died'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>Survived</th>\n",
" <th>Died</th>\n",
" <th>OR</th>\n",
" <th>Interval</th>\n",
" <th>N</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>male</th>\n",
" <td> 0.60</td>\n",
" <td> 0.43</td>\n",
" <td> 0.51</td>\n",
" <td> [0.05, 2.53]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>under 2 months</th>\n",
" <td> 0.20</td>\n",
" <td> 0.43</td>\n",
" <td> 3.00</td>\n",
" <td> [0.27, 15.16]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2-11 months</th>\n",
" <td> 0.59</td>\n",
" <td> 0.29</td>\n",
" <td> 0.28</td>\n",
" <td> [-0.03, 1.15]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12-23 months</th>\n",
" <td> 0.12</td>\n",
" <td> 0.14</td>\n",
" <td> 1.26</td>\n",
" <td> [-0.82, 5.06]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Jordanian</th>\n",
" <td> 0.90</td>\n",
" <td> 0.71</td>\n",
" <td> 0.25</td>\n",
" <td> [-2.08, 3.17]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Palestinian</th>\n",
" <td> 0.07</td>\n",
" <td> 0.29</td>\n",
" <td> 5.51</td>\n",
" <td> [-0.7, 23.23]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 20</th>\n",
" <td> 0.61</td>\n",
" <td> 0.71</td>\n",
" <td> 1.46</td>\n",
" <td> [-12.71, 17.4]</td>\n",
" <td> (1186, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>vitamin D &lt; 11</th>\n",
" <td> 0.45</td>\n",
" <td> 0.43</td>\n",
" <td> 0.93</td>\n",
" <td> [0.08, 4.79]</td>\n",
" <td> (1186, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>prev_cond</th>\n",
" <td> 0.07</td>\n",
" <td> 0.29</td>\n",
" <td> 5.19</td>\n",
" <td> [-0.68, 20.98]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>heart_hx</th>\n",
" <td> 0.03</td>\n",
" <td> 0.00</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>breastfed</th>\n",
" <td> 0.63</td>\n",
" <td> 0.57</td>\n",
" <td> 0.77</td>\n",
" <td> [0.11, 5.85]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>premature</th>\n",
" <td> 0.13</td>\n",
" <td> 0.29</td>\n",
" <td> 2.74</td>\n",
" <td> [-0.35, 11.29]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_pneumo</th>\n",
" <td> 0.16</td>\n",
" <td> 0.57</td>\n",
" <td> 6.94</td>\n",
" <td> [1.12, 50.0]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchopneumo</th>\n",
" <td> 0.34</td>\n",
" <td> 0.00</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_sepsis</th>\n",
" <td> 0.18</td>\n",
" <td> 0.43</td>\n",
" <td> 3.55</td>\n",
" <td> [0.26, 18.26]</td>\n",
" <td> (1390, 7)</td>\n",
" </tr>\n",
" <tr>\n",
" <th>adm_bronchiolitis</th>\n",
" <td> 0.27</td>\n",
" <td> 0.00</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" <td> NaN</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>16 rows \u00d7 5 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 111,
"text": [
" Survived Died OR Interval N\n",
"male 0.60 0.43 0.51 [0.05, 2.53] (1390, 7)\n",
"under 2 months 0.20 0.43 3.00 [0.27, 15.16] (1390, 7)\n",
"2-11 months 0.59 0.29 0.28 [-0.03, 1.15] (1390, 7)\n",
"12-23 months 0.12 0.14 1.26 [-0.82, 5.06] (1390, 7)\n",
"Jordanian 0.90 0.71 0.25 [-2.08, 3.17] (1390, 7)\n",
"Palestinian 0.07 0.29 5.51 [-0.7, 23.23] (1390, 7)\n",
"vitamin D < 20 0.61 0.71 1.46 [-12.71, 17.4] (1186, 7)\n",
"vitamin D < 11 0.45 0.43 0.93 [0.08, 4.79] (1186, 7)\n",
"prev_cond 0.07 0.29 5.19 [-0.68, 20.98] (1390, 7)\n",
"heart_hx 0.03 0.00 NaN NaN NaN\n",
"breastfed 0.63 0.57 0.77 [0.11, 5.85] (1390, 7)\n",
"premature 0.13 0.29 2.74 [-0.35, 11.29] (1390, 7)\n",
"adm_pneumo 0.16 0.57 6.94 [1.12, 50.0] (1390, 7)\n",
"adm_bronchopneumo 0.34 0.00 NaN NaN NaN\n",
"adm_sepsis 0.18 0.43 3.55 [0.26, 18.26] (1390, 7)\n",
"adm_bronchiolitis 0.27 0.00 NaN NaN NaN\n",
"\n",
"[16 rows x 5 columns]"
]
}
],
"prompt_number": 111
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.premature.value_counts()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 112,
"text": [
"0 1218\n",
"1 179\n",
"dtype: int64"
]
}
],
"prompt_number": 112
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Vitamin D"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Number of vitamin D records"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.hospitalized_vitamin_d.notnull().sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 113,
"text": [
"2689"
]
}
],
"prompt_number": 113
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Median vitaimin D"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.hospitalized_vitamin_d.median()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 114,
"text": [
"16.52601"
]
}
],
"prompt_number": 114
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion of subjects with vitamin D < 20"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(hospitalized.hospitalized_vitamin_d<20).mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 115,
"text": [
"0.4925844114862733"
]
}
],
"prompt_number": 115
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Create subset of RSV patients"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only = hospitalized[hospitalized.pcr_result___1==1]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 116
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Number of RSV subjects"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.hospitalized_vitamin_d.notnull().sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 117,
"text": [
"1193"
]
}
],
"prompt_number": 117
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Median vitamin D of RSV subset"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.hospitalized_vitamin_d.median()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 118,
"text": [
"14.3"
]
}
],
"prompt_number": 118
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Proportion of RSV subjects with vitamin D < 20"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"(rsv_only.hospitalized_vitamin_d<20).mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 119,
"text": [
"0.51968503937007871"
]
}
],
"prompt_number": 119
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"non_rsv = hospitalized[hospitalized.pcr_result___1==0]"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 120
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"This model estimates the difference in having vitamin D less than 20 for RSV-only versus non-RSV."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"from pymc import Model, Beta, Deterministic, Binomial\n",
"\n",
"n = non_rsv.shape[0]\n",
"n_rsv = rsv_only.shape[0]\n",
"\n",
"with Model() as model:\n",
" \n",
" p = Beta('p', 1, 1)\n",
" p_rsv = Beta('p_rsv', 1, 1)\n",
" \n",
" p_diff = Deterministic('p_diff', p_rsv - p)\n",
" p_less = Deterministic('p_less', p_rsv < p)\n",
" \n",
" y = Binomial('y', n=n, p=p, observed=(non_rsv.hospitalized_vitamin_d<20).sum())\n",
" y_rsv = Binomial('y_rsv', n=n_rsv, p=p_rsv, observed=(rsv_only.hospitalized_vitamin_d<20).sum())\n",
" "
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 121
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"with model:\n",
" \n",
" step = NUTS()\n",
" trace = sample(2000, step)"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"_ = traceplot(trace, vars=[p, p_rsv])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The difference is positive, with an expected value of approximately 5% higher for the RSV subset."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"_ = traceplot(trace, vars=[p_diff])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"summary(trace, vars=[p_diff, p_less])"
],
"language": "python",
"metadata": {},
"outputs": []
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Factors associated with death\n",
"\n",
"Number and proportion died:"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.death.value_counts().rename({False: 'Survived', True: 'Died'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 122,
"text": [
"Survived 3138\n",
"Died 31\n",
"dtype: int64"
]
}
],
"prompt_number": 122
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized.death.mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 123,
"text": [
"0.0097822656989586618"
]
}
],
"prompt_number": 123
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death = hospitalized.groupby('death')"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 124
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Median z-score"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death['z_score'].median().rename({False: 'Survived', True: 'Died'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 125,
"text": [
"death\n",
"Survived 0.01\n",
"Died -2.53\n",
"Name: z_score, dtype: float64"
]
}
],
"prompt_number": 125
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Gestational age"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death['gest_age'].median().rename({False: 'Survived', True: 'Died'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 126,
"text": [
"death\n",
"Survived 40\n",
"Died 40\n",
"Name: gest_age, dtype: int64"
]
}
],
"prompt_number": 126
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Vitamin D"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death['hospitalized_vitamin_d'].median().rename({False: 'Survived', True: 'Died'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 127,
"text": [
"death\n",
"Survived 16.5\n",
"Died 20.4\n",
"Name: hospitalized_vitamin_d, dtype: float64"
]
}
],
"prompt_number": 127
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Previous conditions"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death[[c for c in hospitalized.columns if c.endswith('hx') and not c.startswith('no_')]].mean().rename({False: 'Survived', True: 'Died'})"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>diabetes_hx</th>\n",
" <th>heart_hx</th>\n",
" <th>kidney_hx</th>\n",
" <th>downs_hx</th>\n",
" <th>sickle_hx</th>\n",
" <th>cancer_hx</th>\n",
" <th>cf_hx</th>\n",
" <th>genetic_hx</th>\n",
" <th>cp_hx</th>\n",
" <th>seizure_hx</th>\n",
" <th>neuro_hx</th>\n",
" <th>mr_hx</th>\n",
" <th>asthma_hx</th>\n",
" <th>immuno_hx</th>\n",
" <th>liver_hx</th>\n",
" <th>gerd_hx</th>\n",
" <th>diarrhea_hx</th>\n",
" <th>other_hx</th>\n",
" </tr>\n",
" <tr>\n",
" <th>death</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>Survived</th>\n",
" <td> 0.001912</td>\n",
" <td> 0.045252</td>\n",
" <td> 0.003824</td>\n",
" <td> 0.012747</td>\n",
" <td> 0.000319</td>\n",
" <td> 0.000319</td>\n",
" <td> 0.000637</td>\n",
" <td> 0.005736</td>\n",
" <td> 0.003824</td>\n",
" <td> 0.003505</td>\n",
" <td> 0.008604</td>\n",
" <td> 0.000637</td>\n",
" <td> 0.007967</td>\n",
" <td> 0.000319</td>\n",
" <td> 0.001593</td>\n",
" <td> 0.000637</td>\n",
" <td> 0</td>\n",
" <td> 0.019758</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Died</th>\n",
" <td> 0.000000</td>\n",
" <td> 0.129032</td>\n",
" <td> 0.032258</td>\n",
" <td> 0.064516</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.096774</td>\n",
" <td> 0.064516</td>\n",
" <td> 0.032258</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0</td>\n",
" <td> 0.096774</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>2 rows \u00d7 18 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 128,
"text": [
" diabetes_hx heart_hx kidney_hx downs_hx sickle_hx cancer_hx \\\n",
"death \n",
"Survived 0.001912 0.045252 0.003824 0.012747 0.000319 0.000319 \n",
"Died 0.000000 0.129032 0.032258 0.064516 0.000000 0.000000 \n",
"\n",
" cf_hx genetic_hx cp_hx seizure_hx neuro_hx mr_hx \\\n",
"death \n",
"Survived 0.000637 0.005736 0.003824 0.003505 0.008604 0.000637 \n",
"Died 0.000000 0.096774 0.064516 0.032258 0.000000 0.000000 \n",
"\n",
" asthma_hx immuno_hx liver_hx gerd_hx diarrhea_hx other_hx \n",
"death \n",
"Survived 0.007967 0.000319 0.001593 0.000637 0 0.019758 \n",
"Died 0.000000 0.000000 0.000000 0.000000 0 0.096774 \n",
"\n",
"[2 rows x 18 columns]"
]
}
],
"prompt_number": 128
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Organism positive culture"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"groupby_death[pcr_lookup.keys()].mean().rename(columns=pcr_lookup)"
],
"language": "python",
"metadata": {},
"outputs": [
{
"html": [
"<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>flu B</th>\n",
" <th>rhino</th>\n",
" <th>PIV1</th>\n",
" <th>PIV2</th>\n",
" <th>RSV</th>\n",
" <th>HMPV</th>\n",
" <th>flu A</th>\n",
" <th>PIV3</th>\n",
" <th>Adeno</th>\n",
" <th>Swine H1</th>\n",
" <th>flu C</th>\n",
" <th>H3N2</th>\n",
" <th>Swine</th>\n",
" <th>H1N1</th>\n",
" </tr>\n",
" <tr>\n",
" <th>death</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>False</th>\n",
" <td> 0.009242</td>\n",
" <td> 0.390376</td>\n",
" <td> 0.010835</td>\n",
" <td> 0.004143</td>\n",
" <td> 0.442957</td>\n",
" <td> 0.086361</td>\n",
" <td> 0.022945</td>\n",
" <td> 0.04079</td>\n",
" <td> 0.150414</td>\n",
" <td> 0.010835</td>\n",
" <td> 0.006055</td>\n",
" <td> 0.009879</td>\n",
" <td> 0.010198</td>\n",
" <td> 0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>True </th>\n",
" <td> 0.000000</td>\n",
" <td> 0.419355</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.225806</td>\n",
" <td> 0.064516</td>\n",
" <td> 0.129032</td>\n",
" <td> 0.00000</td>\n",
" <td> 0.096774</td>\n",
" <td> 0.032258</td>\n",
" <td> 0.000000</td>\n",
" <td> 0.064516</td>\n",
" <td> 0.032258</td>\n",
" <td> 0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>2 rows \u00d7 14 columns</p>\n",
"</div>"
],
"metadata": {},
"output_type": "pyout",
"prompt_number": 129,
"text": [
" flu B rhino PIV1 PIV2 RSV HMPV flu A \\\n",
"death \n",
"False 0.009242 0.390376 0.010835 0.004143 0.442957 0.086361 0.022945 \n",
"True 0.000000 0.419355 0.000000 0.000000 0.225806 0.064516 0.129032 \n",
"\n",
" PIV3 Adeno Swine H1 flu C H3N2 Swine H1N1 \n",
"death \n",
"False 0.04079 0.150414 0.010835 0.006055 0.009879 0.010198 0 \n",
"True 0.00000 0.096774 0.032258 0.000000 0.064516 0.032258 0 \n",
"\n",
"[2 rows x 14 columns]"
]
}
],
"prompt_number": 129
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"direct to icu versus during hospitalization?? compare ourcome such as survival, days on the ventilator, duration of hospitalization and age group as well as z score\n",
"\n",
"any predictors on those who were first admitted to the floor and then transferred, namely should they have been sent there in the first place??\n",
"\n",
"what is the difference in the age, sex and diagnosis of the two groups and in comparison with the total group\n",
"\n",
"what about the blood culture positivity in the two groups and versus the whole cohort\n",
"\n",
"duration of hospitalization in the two groups and the whole cohort\n",
"\n",
"comparison of these with the whole cohort is needed in order to determine the following from our study\n",
"\n",
"- why are some kids sicker than others?\n",
"- what are the characteristics that a doctor should look for (predictors) in order to send them directly to the ICU\n",
"- such characterisitcs should include, age, premature status, history of prior hospitalization ( if we have it) history of chronic disease or other medical conditions, history of having been on bronchodilators\n",
"- younger age\n",
"- low birth weight\n",
"- and low z score (look at less than one and less than two)\n",
"- duration of illness prior to admission\n",
"- cyanosis\n",
"- low O2 saturation on admission\n",
"- septic looking?? dont know if we have that on record\n",
"- vitamin d values\n",
"- actual organism ( virus )\n",
"- viral load\n",
"- antibiotic used especialy when compared with organism cultured in case of positive blood culture\n",
"- medical diagnosis, pneumonia, bronchopneumonia etc"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## ICU by diagnosis"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized[hospitalized.icu_any==1]['adm_pneumo'].sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 130,
"text": [
"82"
]
}
],
"prompt_number": 130
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized[hospitalized.icu_any==1]['adm_bronchopneumo'].sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 131,
"text": [
"66"
]
}
],
"prompt_number": 131
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"hospitalized[hospitalized.icu_any==1]['adm_bronchiolitis'].sum()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 132,
"text": [
"37"
]
}
],
"prompt_number": 132
},
{
"cell_type": "heading",
"level": 3,
"metadata": {},
"source": [
"Comparison of vitamin D levels"
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.groupby('oxygen')['hospitalized_vitamin_d'].hist()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 152,
"text": [
"oxygen\n",
"0 Axes(0.125,0.125;0.775x0.775)\n",
"1 Axes(0.125,0.125;0.775x0.775)\n",
"Name: hospitalized_vitamin_d, dtype: object"
]
},
{
"metadata": {},
"output_type": "display_data",
"png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAECCAYAAADw0Rw8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text": [
"<matplotlib.figure.Figure at 0x114ef6710>"
]
}
],
"prompt_number": 152
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.groupby('oxygen')['hospitalized_vitamin_d'].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 126,
"text": [
"oxygen\n",
"0 16.528299\n",
"1 14.086796\n",
"Name: hospitalized_vitamin_d, dtype: float64"
]
}
],
"prompt_number": 126
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only['los5'] = (rsv_only.length_of_stay>=5).astype(int)"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 129
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.groupby('los5')['hospitalized_vitamin_d'].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 131,
"text": [
"los5\n",
"0 16.505338\n",
"1 14.688211\n",
"Name: hospitalized_vitamin_d, dtype: float64"
]
}
],
"prompt_number": 131
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.groupby('adm_pneumo')['hospitalized_vitamin_d'].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 132,
"text": [
"adm_pneumo\n",
"0 16.250651\n",
"1 11.975369\n",
"Name: hospitalized_vitamin_d, dtype: float64"
]
}
],
"prompt_number": 132
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The following calculates the estimated median difference between the means of each group, along with a 95% bootstrap confidence interval. These can be interpreted as the difference in vitamin D levels between the two groups, and it is robust to the fact that vitamin D levels are not normally distributed within groups."
]
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"def bootstrap_difference(df, var, group, alpha=0.05, samples=10000):\n",
" \n",
" grouped_var = df.groupby(group)[var]\n",
" diffs = [np.diff(grouped_var.apply(lambda x: x.iloc[np.random.randint(len(x), size=len(x))].mean())) for i in range(samples)]\n",
" \n",
" interval = np.percentile(diffs, [100*(alpha/2.), 100*(1. - alpha/2.)])\n",
" \n",
" return np.round(np.median(diffs), 2), np.round(interval, 2).tolist()"
],
"language": "python",
"metadata": {},
"outputs": [],
"prompt_number": 148
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bootstrap_difference(rsv_only, 'hospitalized_vitamin_d', 'adm_pneumo')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 149,
"text": [
"(-4.29, [-6.06, -2.46])"
]
}
],
"prompt_number": 149
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bootstrap_difference(rsv_only, 'hospitalized_vitamin_d', 'adm_bronchiolitis')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 153,
"text": [
"(0.050000000000000003, [-1.59, 1.71])"
]
}
],
"prompt_number": 153
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bootstrap_difference(rsv_only, 'hospitalized_vitamin_d', 'adm_bronchopneumo')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 154,
"text": [
"(4.8499999999999996, [3.39, 6.32])"
]
}
],
"prompt_number": 154
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bootstrap_difference(rsv_only, 'hospitalized_vitamin_d', 'adm_sepsis')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 155,
"text": [
"(-4.6500000000000004, [-6.34, -2.88])"
]
}
],
"prompt_number": 155
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"rsv_only.groupby('adm_sepsis')['hospitalized_vitamin_d'].mean()"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 156,
"text": [
"adm_sepsis\n",
"0 16.332006\n",
"1 11.701480\n",
"Name: hospitalized_vitamin_d, dtype: float64"
]
}
],
"prompt_number": 156
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bootstrap_difference(rsv_only, 'hospitalized_vitamin_d', 'oxygen')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 150,
"text": [
"(-2.4300000000000002, [-3.87, -1.0])"
]
}
],
"prompt_number": 150
},
{
"cell_type": "code",
"collapsed": false,
"input": [
"bootstrap_difference(rsv_only, 'hospitalized_vitamin_d', 'los5')"
],
"language": "python",
"metadata": {},
"outputs": [
{
"metadata": {},
"output_type": "pyout",
"prompt_number": 151,
"text": [
"(-1.8200000000000001, [-3.24, -0.39])"
]
}
],
"prompt_number": 151
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Thus, none of the intervals above include zero, making it unlikely that the underlying populations are the same."
]
}
],
"metadata": {}
}
]
}
@sburns
Copy link

sburns commented Feb 17, 2014

PHI in the early hospitalized.head()?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment