Skip to content

Instantly share code, notes, and snippets.

@fonnesbeck

fonnesbeck/UF Interventions Meta-analysis (PyMC 2).ipynb

Created February 4, 2016 21:12
Show Gist options
  • Save fonnesbeck/ea8bae8b6686c213920a to your computer and use it in GitHub Desktop.
Save fonnesbeck/ea8bae8b6686c213920a to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
UF Interventions Meta-analysis (PyMC 2).ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Uterine fibroids follow-up treatment meta-analysis\n",
"\n",
"Our goal is to estimate the probabilities of requiring one of a suite of candidate follow-up treatments following randomization to a given initial treatment for uterine fibroids. Specifically, we are interested in estimating:\n",
"\n",
"$$Pr(I_2|I_1 =i,T=t)$$\n",
"\n",
"where $I_1$ is an initial intervention, which take specific values $i = 1, 2, \\ldots , K$ for each of $K$ candidate intervention types, $I_2$ is the followup intervention that also may take any of the same values of $i$, and $T$ is followup time in months, which will generally be either 6 or 12 months.\n",
"\n",
"Our current set of candidate interventions include:\n",
"\n",
"- Myomectomy\n",
"- Hysterectomy\n",
"- Ablation\n",
"- UAE\n",
"- Magnetic resonance imaging-guided high-intensity focused ultrasound (MRIgFUS) \n",
"- Ablation +/- hysteroscopic myomectomy\n",
"- No intervention\n",
"\n",
"Rather than model each conditional probability independently, we will instead model the outcomes for a treatment arm as a multinomial random variable. That is,\n",
"\n",
"$$\\{X_{I_2} \\} ∼ \\text{Multinomial}(N_{I_1}=i, \\{\\pi_i\\})$$\n",
"\n",
"where $\\{X_{I_2}\\}$ is the vector of outcomes corresponding to each of the possible followup interventions listed above, $N_{I_1}=i$ is the number of women randomized to the initial intervention i, and $\\{\\pi_i\\}$ is a vector of conditional transition probabilities corresponding to $Pr(I_2|I_1 = i, T = t)$, as specified above. The multinomial distribution is a multivariate generalization of the categorical distribution, which is what the above simplifies to when modeling the outcome for a single patient. The multivariate formulation allows us to model study-arm-specific outcomes, incorporating covariates that are specific to that arm or study.\n",
" \n",
"The quantities of interest are the vectors of transition probabilities $\\{\\pi_i\\}$ corresponding to each of the initial candidate interventions. A naive approach to modeling these is to assign a vague Dirichlet prior distribution to each set, and perform Bayesian inference using the multinomial likelihood, with which the Dirichlet is conjugate, to yield posterior estimates for each probability. However, there may be additional information with which to model these probabilities, which may include:\n",
"\n",
"- followup time for each study\n",
"- arm-specific demographic covariates (e.g. race, mean age) \n",
"- study-specific random effects\n",
"\n",
"hence, a given transition probability $\\pi_{ijk}$ – the probability of transitioning from initial intervention $i$ to followup intervention $j$ in study $k$ – may be modeled as:\n",
"\n",
"$$\\text{logit}(\\pi_{ijk})= \\theta_{ij} + X_k \\beta_{ij} + \\epsilon_k$$\n",
"\n",
"where $\\theta_{ij}$ is a baseline transition probability (on the logit scale), $X_k$ a matrix of study(-arm)-specific covariates, $\\beta_{ij}$ the corresponding coefficients, and $\\epsilon_k$ a mean-zero random effect for study k. We will initially consider (1) follow-up time and (2) mean/median age as covariates.\n",
" \n",
"An attractive benefit to using Bayesian inference to estimate this model is that it is easy to generate predictions from the model, via the posterior predictive distribution. For example, we could estimate the distribution of the expected proportion of women requiring a particular followup intervention; this estimate would factor in both the residual uncertainty in the transition probability estimates, as well as the sampling uncertainty of the intervention."
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"%matplotlib inline\n",
"import numpy as np\n",
"import pandas as pd\n",
"import pymc as pm\n",
"import seaborn as sns\n",
"import pdb\n",
"sns.set()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Import data from worksheets in Excel spreadsheet."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"data_file = 'UF Subsequent Interventions Data_Master_updated.xlsx'"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Occlusion rows=31, columns=13, missing=6\n",
"Med vs IAC rows=49, columns=13, missing=46\n",
"Med vs Med rows=67, columns=13, missing=13\n",
"UAE rows=32, columns=13, missing=0\n"
]
}
],
"source": [
"missing = ['NA', 'NR', 'ND', '?', 'null']\n",
"\n",
"misc_data = pd.read_excel('data/' + data_file, sheetname='MISC (SP)', na_values=missing)\n",
"misc_data = misc_data[~misc_data['baseline_n'].isnull()].drop('notes', axis=1)\n",
"rows, cols = misc_data.shape\n",
"print('Occlusion rows={0}, columns={1}, missing={2}'.format(rows, cols,\n",
" misc_data.isnull().sum().sum()))\n",
"\n",
"med_vs_iac_data = pd.read_excel('data/' + data_file, sheetname='Med vs IAC JW', na_values=missing)\n",
"med_vs_iac_data = med_vs_iac_data[~med_vs_iac_data['trial_arm'].isnull()].drop('notes', axis=1)\n",
"rows, cols = med_vs_iac_data.shape\n",
"print('Med vs IAC rows={0}, columns={1}, missing={2}'.format(rows, cols, \n",
" med_vs_iac_data.isnull().sum().sum()))\n",
"\n",
"med_vs_med_data = pd.read_excel('data/' + data_file, sheetname='Med vs Med DVE', na_values=missing)\n",
"med_vs_med_data = med_vs_med_data[~med_vs_med_data['baseline_n'].isnull()].drop('notes', axis=1)\n",
"rows, cols = med_vs_med_data.shape\n",
"print('Med vs Med rows={0}, columns={1}, missing={2}'.format(rows, cols, \n",
" med_vs_med_data.isnull().sum().sum()))\n",
"\n",
"uae_data = pd.read_excel('data/' + data_file, sheetname='UAE SK')\n",
"uae_data = uae_data[~uae_data['baseline_n'].isnull()].drop('notes', axis=1)\n",
"rows, cols = uae_data.shape\n",
"print('UAE rows={0}, columns={1}, missing={2}'.format(rows, cols, \n",
" uae_data.isnull().sum().sum()))\n",
"\n",
"datasets = [misc_data, med_vs_iac_data, med_vs_med_data, uae_data]"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"unique_inerventions = set(np.concatenate([d.intervention.values for d in datasets]))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Use the following lookup table to create \"intervention category\" field in each dataset."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"# %load intervention_lookup.py\n",
"intervention_lookup = {'Ablation': 'ablation',\n",
" 'Ablation+/- hysteroscopic myomectomy': 'ablation',\n",
" 'Asoprisnil 10 mg': 'med_manage',\n",
" 'Asoprisnil 25 mg': 'med_manage',\n",
" 'Asoprisnil 5 mg': 'med_manage',\n",
" 'CD20 (Ulipristal)': 'med_manage',\n",
" 'CDB10 (Ulipristal)': 'med_manage',\n",
" 'Hysterectomy': 'hysterectomy',\n",
" 'LBCUV': 'uae',\n",
" 'LP + GnRH agonist plus raloxifene': 'med_manage',\n",
" 'LP + placebo': 'med_manage',\n",
" 'LPA+ MPA / LPA+placebo': 'med_manage',\n",
" 'LPA+ placebo / LPA+MPA': 'med_manage',\n",
" 'LUNA plus LBCUV': 'ablation',\n",
" 'Myomectomy': 'myomectomy',\n",
" 'No treatment': 'control',\n",
" 'No treatment (control)': 'control',\n",
" 'Placebo': 'control',\n",
" 'Raloxifene, 180mg/day': 'med_manage',\n",
" 'SC implant of 3.6 goserelin + placebo (3 months) then tibolone 2.5 mg daily (3 months)': 'med_manage',\n",
" 'SC implant of 3.6 goserelin + placebo (6 months)': 'med_manage',\n",
" 'SC implant of 3.6 goserelin + tibolone 2.5 mg daily (6 months)': 'med_manage',\n",
" 'Surgery': 'DROP',\n",
" 'Tibolone': 'med_manage',\n",
" 'UAE': 'uae',\n",
" 'UAE only': 'uae',\n",
" 'UAE plus goserelin acetate depot': 'uae',\n",
" 'buserelin + goserelin': 'med_manage',\n",
" 'buserelin, intranasal': 'med_manage',\n",
" 'cabergoline': 'med_manage',\n",
" 'diphereline': 'med_manage',\n",
" 'gestrinone, 2.5mg': 'med_manage',\n",
" 'gestrinone, 2.5mg oral + gestrinone, 5mg oral + gestrinone, 5mg vaginal': 'med_manage',\n",
" 'gestrinone, 5mg': 'med_manage',\n",
" 'gestrinone, 5mg vaginal': 'med_manage',\n",
" 'goserelin, subcutaneous': 'med_manage',\n",
" 'healthy controls': 'control',\n",
" 'hormone replacement therapy, transdermal': 'DROP',\n",
" 'hysterectomy or myomectomy': 'DROP',\n",
" 'letrozole, 2.5mg': 'med_manage',\n",
" 'leuprolide': 'med_manage',\n",
" 'leuprolide acetate depot (11.25 mg q 3 months) + Placebo': 'med_manage',\n",
" 'leuprolide acetate depot (11.25 mg q 3 months) + tibolone 2.5 mg/d orally': 'med_manage',\n",
" 'leuprolide acetate depot (3.75 mg/28 d) + placebo (B)': 'med_manage',\n",
" 'leuprolide plus (tibolone 2.5 mg daily) (A)': 'med_manage',\n",
" 'leuprolide plus MPA': 'med_manage',\n",
" 'leuprolide plus estrogen-progestin': 'med_manage',\n",
" 'leuprolide plus placebo': 'med_manage',\n",
" 'leuprolide plus progestin': 'med_manage',\n",
" 'leuprolide plus raloxifene 60 mg daily': 'med_manage',\n",
" 'leuprolide, 1.88mg': 'med_manage',\n",
" 'leuprolide, 3.75mg': 'med_manage',\n",
" 'mifepristone, 10mg': 'med_manage',\n",
" 'mifepristone, 10mg + mifepristone, 5mg': 'med_manage',\n",
" 'mifepristone, 2.5mg': 'med_manage',\n",
" 'mifepristone, 5mg': 'med_manage',\n",
" 'placebo': 'control',\n",
" 'raloxifene 180 mg daily': 'med_manage',\n",
" 'raloxifene 60 mg daily': 'med_manage',\n",
" 'tamoxifen 20 mg daily': 'med_manage',\n",
" 'tibolone': 'med_manage',\n",
" 'tibolone, 2.5mg': 'med_manage',\n",
" 'transdermal estrogen replacement therapy': 'med_manage',\n",
" 'triptorelin, 100ug': 'med_manage',\n",
" 'triptorelin, 100ug + triptorelin, 20ug + triptorelin, 5ug': 'med_manage',\n",
" 'triptorelin, 20ug': 'med_manage',\n",
" 'triptorelin, 3.6mg/mo': 'med_manage',\n",
" 'triptorelin, 5ug': 'med_manage',\n",
" 'ulipristal acetate followed by placebo': 'med_manage',\n",
" 'ulipristal acetate followed by progestin': 'med_manage',\n",
" 'ulipristal, 10mg': 'med_manage',\n",
" 'ulipristal, 5mg': 'med_manage',\n",
" 'HIFU': 'MRgFUS',\n",
" 'HIFU with CEUS': 'MRgFUS',\n",
" 'LUAO': 'uae',\n",
" 'UAE plus PVA': 'uae',\n",
" 'UAE plus TAG': 'uae',\n",
" 'UAE with PVA': 'uae',\n",
" 'UAE with PVA particles, large': 'uae',\n",
" 'UAE with PVA particles, small': 'uae',\n",
" 'UAE with SPA': 'uae',\n",
" 'UAE with SPVA': 'uae',\n",
" 'UAE with TAG': 'uae',\n",
" 'UAE with TAG microspheres': 'uae',\n",
" 'myomectomy': 'myomectomy',\n",
" 'myomectomy with vasopressin': 'myomectomy',\n",
" 'myomectomy, abdominal': 'myomectomy',\n",
" 'myomectomy, laparoscopic': 'myomectomy',\n",
" 'myomectomy, loop ligation with vasopressin': 'myomectomy',\n",
" 'myomectomy, minilaparotomic': 'myomectomy'}\n"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Assign intervention **categories** to each arm"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"datasets = [d.assign(intervention_cat=d.intervention.replace(intervention_lookup)) for d in datasets]"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"{'DROP',\n",
" 'MRgFUS',\n",
" 'ablation',\n",
" 'control',\n",
" 'hysterectomy',\n",
" 'med_manage',\n",
" 'myomectomy',\n",
" 'uae'}"
]
},
"execution_count": 7,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"intervention_categories = set(intervention_lookup.values())\n",
"intervention_categories"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Import demographic information"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"Index(['study_id', 'Citation', 'FamCode', 'FamDesig', 'NCT', 'ArmsN',\n",
" 'ArmCategory', 'Group_Desc', 'New Grouping', 'Demo_Category',\n",
" 'Demo_specify', 'BL N', 'Denom_N', 'BL %', 'BL Mean', 'BL SD', 'BL_SE',\n",
" 'BL_Median', 'BL Min', 'BL Max', 'BL 95% L', 'BL 95% H',\n",
" 'BL_group_diff', 'Comments'],\n",
" dtype='object')"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"demographics = pd.read_excel('data/' + data_file, sheetname='ALL_DEMO_DATA', na_values=missing)\n",
"demographics.columns"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Extract columns of interest"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"age_data = demographics.loc[demographics.Demo_Category=='Age', ['study_id', 'New Grouping', 'BL Mean', 'BL SD']]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Clean arm labels"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"age_data = age_data.assign(arm=age_data['New Grouping'].str.replace(':','')).drop('New Grouping', axis=1)"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array(['G2', 'G1', 'G1b', 'G1a', 'G3', 'CG', 'G1c', 'G1+G2', 'G1a+G1b+G1c'], dtype=object)"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"age_data.arm.unique()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Concatenate all datasets"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"all_data = pd.concat(datasets)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Clean up study arm field"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"all_arm = all_data.trial_arm.str.replace(':','').str.replace(' ', '').str.replace('Group', 'G')\n",
"all_data = all_data.assign(arm=all_arm).drop('trial_arm', axis=1)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array(['G1', 'G2', 'G3', 'CG', 'G1a', 'G1b', 'G1c', 'CG1', 'CG2', 'G1/CG',\n",
" 'CG/G1', 'G1a+G1b', 'G1a+G1b+G1c', 'G1+G2'], dtype=object)"
]
},
"execution_count": 14,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_data.arm.unique()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Clean up study ID field. Currently contains non-numeric entries. Will strip out the first study ID from the compund labels, as this is the parent study ID."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([23, 347, 1400, 1529, 1806, 1889, 2375, 2967, 3382, 3690, 3785, 5186,\n",
" 5474, 414, 1849, 3016, 3181, 3324, 3674, 4258, 4468, 4858, 4960,\n",
" 5276, 5302, 6091, 6263, 6696, 7155, 7504, 7797, 7936, 95.0, 629.0,\n",
" 757.0, 1290.0, 2318.0, 2555.0, 2635.0, 3312.0, 3978.0, 4787.0,\n",
" 4961.0, 5721.0, 6393.0, 6903.0, 7139.0, 7309.0, 7530.0, 7589.0,\n",
" 7763.0, '3803_3052', 1546, '3365_2026_1657_986',\n",
" '3819_815_1986_2759_2971_\\n3120_3175_3192_3678_3721', 4789, 2006], dtype=object)"
]
},
"execution_count": 15,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_data.study_id.unique()"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"str_mask = all_data.study_id.str.isnumeric()==False\n",
"all_data.loc[str_mask, 'study_id'] = all_data.study_id[str_mask].apply(lambda x: x[:x.find('_')])\n",
"all_data.study_id = all_data.study_id.astype(int)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([ 23, 347, 1400, 1529, 1806, 1889, 2375, 2967, 3382, 3690, 3785,\n",
" 5186, 5474, 414, 1849, 3016, 3181, 3324, 3674, 4258, 4468, 4858,\n",
" 4960, 5276, 5302, 6091, 6263, 6696, 7155, 7504, 7797, 7936, 95,\n",
" 629, 757, 1290, 2318, 2555, 2635, 3312, 3978, 4787, 4961, 5721,\n",
" 6393, 6903, 7139, 7309, 7530, 7589, 7763, 3803, 1546, 3365, 3819,\n",
" 4789, 2006])"
]
},
"execution_count": 17,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_data.study_id.unique()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Here is what the data look like after merging."
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>study_id</th>\n",
" <th>intervention</th>\n",
" <th>baseline_n</th>\n",
" <th>followup_interval</th>\n",
" <th>followup_n</th>\n",
" <th>hysterectomy</th>\n",
" <th>myomectomy</th>\n",
" <th>uae</th>\n",
" <th>MRIgFUS</th>\n",
" <th>ablation</th>\n",
" <th>iud</th>\n",
" <th>no_treatment</th>\n",
" <th>intervention_cat</th>\n",
" <th>arm</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>23</td>\n",
" <td>HIFU with CEUS</td>\n",
" <td>17</td>\n",
" <td>12</td>\n",
" <td>17</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>16</td>\n",
" <td>MRgFUS</td>\n",
" <td>G1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>23</td>\n",
" <td>HIFU</td>\n",
" <td>16</td>\n",
" <td>12</td>\n",
" <td>16</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>13</td>\n",
" <td>MRgFUS</td>\n",
" <td>G2</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>347</td>\n",
" <td>UAE with SPVA</td>\n",
" <td>30</td>\n",
" <td>12</td>\n",
" <td>27</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>26</td>\n",
" <td>uae</td>\n",
" <td>G1</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>347</td>\n",
" <td>UAE with TAG</td>\n",
" <td>30</td>\n",
" <td>12</td>\n",
" <td>29</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>29</td>\n",
" <td>uae</td>\n",
" <td>G2</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1400</td>\n",
" <td>UAE</td>\n",
" <td>63</td>\n",
" <td>6</td>\n",
" <td>62</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>56</td>\n",
" <td>uae</td>\n",
" <td>G1</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" study_id intervention baseline_n followup_interval followup_n \\\n",
"0 23 HIFU with CEUS 17 12 17 \n",
"1 23 HIFU 16 12 16 \n",
"2 347 UAE with SPVA 30 12 27 \n",
"3 347 UAE with TAG 30 12 29 \n",
"4 1400 UAE 63 6 62 \n",
"\n",
" hysterectomy myomectomy uae MRIgFUS ablation iud no_treatment \\\n",
"0 0 0 0 1 0 0 16 \n",
"1 0 0 0 3 0 0 13 \n",
"2 1 0 0 0 0 0 26 \n",
"3 0 0 0 0 0 0 29 \n",
"4 0 1 5 0 0 0 56 \n",
"\n",
" intervention_cat arm \n",
"0 MRgFUS G1 \n",
"1 MRgFUS G2 \n",
"2 uae G1 \n",
"3 uae G2 \n",
"4 uae G1 "
]
},
"execution_count": 18,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_data.head()"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"collapsed": false,
"scrolled": true
},
"outputs": [
{
"data": {
"text/plain": [
"intervention_cat\n",
"DROP 8\n",
"MRgFUS 2\n",
"ablation 3\n",
"control 11\n",
"hysterectomy 7\n",
"med_manage 100\n",
"myomectomy 14\n",
"uae 34\n",
"Name: study_id, dtype: int64"
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_data.groupby('intervention_cat')['study_id'].count()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Merge age data with outcomes"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"all_data_merged = pd.merge(all_data, age_data, on=['study_id', 'arm'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"For now, drop arms with no reported followup time (we may want to impute these):"
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"all_data_merged = all_data_merged.dropna(subset=['followup_interval'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Parse followup intervals that are ranges, creating `fup_min` and `fup_max` fields."
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/ipykernel/__main__.py:1: FutureWarning: convert_objects is deprecated. Use the data-type specific converters pd.to_datetime, pd.to_timedelta and pd.to_numeric.\n",
" if __name__ == '__main__':\n"
]
}
],
"source": [
"dataset = all_data_merged.assign(fup_min=0, fup_max=all_data.followup_interval.convert_objects(convert_numeric=True).max()+1)\n",
"range_index = dataset.followup_interval.str.contains('to').notnull()\n",
"range_vals = dataset[range_index].followup_interval.apply(lambda x: x.split(' '))\n",
"dataset.loc[range_index, ['fup_min']] = range_vals.apply(lambda x: float(x[0]))\n",
"dataset.loc[range_index, ['fup_max']] = range_vals.apply(lambda x: float(x[-1]))\n",
"dataset.loc[range_index, ['followup_interval']] = 30.33\n",
"dataset['followup_interval'] = dataset.followup_interval.astype(float)"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>study_id</th>\n",
" <th>intervention</th>\n",
" <th>baseline_n</th>\n",
" <th>followup_interval</th>\n",
" <th>followup_n</th>\n",
" <th>hysterectomy</th>\n",
" <th>myomectomy</th>\n",
" <th>uae</th>\n",
" <th>MRIgFUS</th>\n",
" <th>ablation</th>\n",
" <th>iud</th>\n",
" <th>no_treatment</th>\n",
" <th>intervention_cat</th>\n",
" <th>arm</th>\n",
" <th>BL Mean</th>\n",
" <th>BL SD</th>\n",
" <th>fup_max</th>\n",
" <th>fup_min</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>23</td>\n",
" <td>HIFU with CEUS</td>\n",
" <td>17</td>\n",
" <td>12</td>\n",
" <td>17</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>16</td>\n",
" <td>MRgFUS</td>\n",
" <td>G1</td>\n",
" <td>43.1</td>\n",
" <td>5.3</td>\n",
" <td>61</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>23</td>\n",
" <td>HIFU</td>\n",
" <td>16</td>\n",
" <td>12</td>\n",
" <td>16</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>3</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>13</td>\n",
" <td>MRgFUS</td>\n",
" <td>G2</td>\n",
" <td>42</td>\n",
" <td>5.4</td>\n",
" <td>61</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>347</td>\n",
" <td>UAE with SPVA</td>\n",
" <td>30</td>\n",
" <td>12</td>\n",
" <td>27</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>26</td>\n",
" <td>uae</td>\n",
" <td>G1</td>\n",
" <td>43.9</td>\n",
" <td>5.0</td>\n",
" <td>61</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>347</td>\n",
" <td>UAE with TAG</td>\n",
" <td>30</td>\n",
" <td>12</td>\n",
" <td>29</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>29</td>\n",
" <td>uae</td>\n",
" <td>G2</td>\n",
" <td>41.7</td>\n",
" <td>5.4</td>\n",
" <td>61</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>1400</td>\n",
" <td>UAE</td>\n",
" <td>63</td>\n",
" <td>6</td>\n",
" <td>62</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>5</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>56</td>\n",
" <td>uae</td>\n",
" <td>G1</td>\n",
" <td>41</td>\n",
" <td>3.5</td>\n",
" <td>61</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" study_id intervention baseline_n followup_interval followup_n \\\n",
"0 23 HIFU with CEUS 17 12 17 \n",
"1 23 HIFU 16 12 16 \n",
"2 347 UAE with SPVA 30 12 27 \n",
"3 347 UAE with TAG 30 12 29 \n",
"4 1400 UAE 63 6 62 \n",
"\n",
" hysterectomy myomectomy uae MRIgFUS ablation iud no_treatment \\\n",
"0 0 0 0 1 0 0 16 \n",
"1 0 0 0 3 0 0 13 \n",
"2 1 0 0 0 0 0 26 \n",
"3 0 0 0 0 0 0 29 \n",
"4 0 1 5 0 0 0 56 \n",
"\n",
" intervention_cat arm BL Mean BL SD fup_max fup_min \n",
"0 MRgFUS G1 43.1 5.3 61 0 \n",
"1 MRgFUS G2 42 5.4 61 0 \n",
"2 uae G1 43.9 5.0 61 0 \n",
"3 uae G2 41.7 5.4 61 0 \n",
"4 uae G1 41 3.5 61 0 "
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dataset.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Fill missing values"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"dataset.loc[dataset.followup_n.isnull(), 'followup_n'] = dataset.loc[dataset.followup_n.isnull(), 'baseline_n']"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"dataset.loc[dataset.no_treatment.isnull(), 'no_treatment'] = dataset.followup_n - dataset[[ 'hysterectomy', 'myomectomy', 'uae',\n",
" 'MRIgFUS', 'ablation', 'iud']].sum(1)[dataset.no_treatment.isnull()]"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([ 12. , 6. , 30.33, 24. , 2. , 1. , 3. , 5.5 ,\n",
" 9. , 18. , 0. , 7. , 60. ])"
]
},
"execution_count": 26,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dataset.followup_interval.unique()"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([43.1, 42, 43.9, 41.7, 41, 43.5, 40.3, 42.7, 45, 44, 38.26, 32.1,\n",
" 34.3, 44.9, 42.5, 43.3, 38.4, 37.5, 45.9, 44.5, 33.97, 34, 41.3,\n",
" 42.9, 42.1, 43.4, 37.7, 43, 40.2, 41.1, 49.1, 48.6, 36.3, 35.9,\n",
" 37.2, 54.2, 51.2, 43.6, nan, 38.9, 37.1, 41.4, 36.9, 41.6, 39, 39.6,\n",
" 39.67, 36.87, 30.94, 31, 39.5, 42.8, 56.2, 57.9, 50.2, 50.6, 34.4,\n",
" 42.2, 49.2, 32.6, 48.4, 33.8, 38.1, 37, 32.3, 43.2, 44.6, 45.4,\n",
" 46.4, 48.5, 48.3], dtype=object)"
]
},
"execution_count": 27,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"dataset['BL Mean'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"crossover_studies = 7155, 3324, 414, 95, 7139, 6903, 3721, 3181, 4858, 4960, 4258, 4789, 2006, 2318"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"outcome_cats = [ 'hysterectomy', 'myomectomy', 'uae',\n",
" 'MRIgFUS', 'ablation', 'iud', 'no_treatment']"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"dataset.loc[dataset['BL Mean'].isnull(), 'BL Mean'] = 90"
]
},
{
"cell_type": "code",
"execution_count": 58,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"from numpy.ma import masked_values\n",
"\n",
"\n",
"def specify_model(intervention):\n",
" \n",
" '''\n",
" Data preparation\n",
" '''\n",
" \n",
" intervention_data = dataset[(dataset.intervention_cat==intervention)\n",
" & ~dataset[outcome_cats].isnull().sum(axis=1).astype(bool)]\n",
" \n",
" followup_masked = masked_values(intervention_data.followup_interval.values, 30.33)\n",
" followup_min, followup_max = intervention_data[['fup_min', 'fup_max']].values.T\n",
"\n",
" outcomes = intervention_data[[ 'hysterectomy', 'myomectomy', 'uae',\n",
" 'MRIgFUS', 'ablation', 'iud', 'no_treatment']].values\n",
" \n",
" if np.isnan(outcomes).any():\n",
" print('Missing values in outcomes for', intervention)\n",
"\n",
" followup_n = intervention_data.followup_n.values\n",
"\n",
" # Center age at 40\n",
" age_masked = masked_values(intervention_data['BL Mean'].values - 40, 50)\n",
" \n",
" studies = intervention_data.study_id.unique()\n",
" study_index = np.array([np.argwhere(studies==i).squeeze() for i in intervention_data.study_id])\n",
"\n",
" study_id = intervention_data.study_id.values\n",
" \n",
" n_studies = len(set(study_id))\n",
"\n",
" n_outcomes = 7\n",
" arms = len(outcomes)\n",
" \n",
" '''\n",
" Model specification\n",
" '''\n",
" \n",
" \n",
" # Impute followup times uniformly over the observed range\n",
" if np.any(followup_masked.mask):\n",
" followup_time = pm.Uniform('followup_time', followup_min.min(), followup_max.max(), \n",
" observed=True, \n",
" value=followup_masked)\n",
" else:\n",
" followup_time = followup_masked.data.astype(float)\n",
"\n",
" # Impute age using a T-distribution\n",
" if np.any(age_masked.mask):\n",
" nu = pm.Exponential('nu', 0.01, value=10)\n",
" age_centered = pm.T('age_centered', nu, observed=True, value=age_masked)\n",
" else:\n",
" age_centered = age_masked.data.astype(float)\n",
"\n",
" # Mean probabilities (on logit scale)\n",
" μ = pm.Normal('μ', 0, 0.01, value=[-1.]*n_outcomes)\n",
" # Followup time covariates \n",
" β_fup = pm.Normal('β_fup', 0, 1e-6, value=[-0.5]*6 + [0])\n",
" # Age covariate\n",
" β_age = pm.Normal('β_age', 0, 1e-6, value=[-0.5]*6 + [0])\n",
"\n",
" # Study random effect\n",
" σ = pm.Uniform('σ', 0, 100, value=1)\n",
" ϵ = pm.Normal('ϵ', 0, σ**-2, value=np.zeros(n_studies))\n",
"\n",
" # Expected value (on logit scale)\n",
" θ_uae = [pm.Lambda('θ_uae_%i' % i, lambda μ=μ, β_fup=β_fup, β_age=β_age, ϵ=ϵ, t=followup_time, a=age_centered: \n",
" np.exp(μ + β_fup*t[i] + β_age*a[i] + ϵ[study_index[i]])) \n",
" for i in range(arms)]\n",
"\n",
" # Inverse-logit transformation to convert to probabilities\n",
" π = [pm.Lambda('π_%i' % i, lambda t=t: t/t.sum()) for i,t in enumerate(θ_uae)]\n",
"\n",
" # Multinomial data likelihood\n",
" y = [pm.Multinomial('y_%i' % i, outcomes[i].sum(), π[i], \n",
" value=outcomes[i], observed=True) for i in range(arms)]\n",
" p_6 = pm.Lambda('p_6', lambda μ=μ, β_fup=β_fup: np.exp(μ + β_fup*6)/np.exp(μ + β_fup*6).sum())\n",
" p_12 = pm.Lambda('p_12', lambda μ=μ, β_fup=β_fup: np.exp(μ + β_fup*12)/np.exp(μ + β_fup*12).sum())\n",
" p_6_50 = pm.Lambda('p_6_50', lambda μ=μ, β_fup=β_fup, β_age=β_age: \n",
" np.exp(μ + β_fup*6 + β_age*10)/np.exp(μ + β_fup*6 + β_age*10).sum())\n",
" \n",
" return locals()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Instantiate models"
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"n_iterations = 50000\n",
"n_burn = 40000\n",
"n_chains = 2"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"chain 1\n",
" [-----------------100%-----------------] 50000 of 50000 complete in 2139.6 sec\n",
"chain 2\n",
" [---------------- 43% ] 21706 of 50000 complete in 9302.4 sec"
]
}
],
"source": [
"uae_model = pm.MCMC(specify_model('uae'))\n",
"for chain in range(n_chains):\n",
" print('\\nchain',chain+1)\n",
" uae_model.sample(n_iterations, n_burn)"
]
},
{
"cell_type": "code",
"execution_count": 41,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"plot_labels = dataset.columns[5:12]"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"array([[ 18.58231743, 28.32509498, 11.52847809, 58.35255279],\n",
" [ 18.58231743, 28.32509498, 11.52847809, 58.35255279],\n",
" [ 18.58231743, 28.32509498, 11.52847809, 58.35255279],\n",
" ..., \n",
" [ 6.41701153, 55.6964252 , 19.74318293, 44.20362134],\n",
" [ 6.41701153, 55.6964252 , 19.74318293, 44.20362134],\n",
" [ 6.41701153, 55.6964252 , 19.74318293, 44.20362134]])"
]
},
"execution_count": 45,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"uae_model.followup_time.trace()"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAfEAAAF7CAYAAADCGbLDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xl4VPWhxvF3JsNiWAwq2Fq5IKBG4MYl4xVZZAuEyDoI\nalGWe7kiClcQK4ugyGZBWlQCkcUqNcqjbQWJ5bFQEAVBlmA1DVzEAgEFxDQ3IJBAMpnf/SPNSCAh\nE0gy5xe/n+fxMZnlnHfOnDnvnCU/XMYYIwAAYB13uAMAAIBLQ4kDAGApShwAAEtR4gAAWIoSBwDA\nUpQ4AACWosThWMnJyerRo4d8Pp+eeuopnThxInhfmzZt5PP5gv/9+c9/liS988476tmzpx588EEd\nPnw4+PgRI0Zo//79F53f559/rv/+7/+Wz+dT7969NXLkSH399deX/TpmzJihBQsWSJIeffRR7du3\nT9u3b1fv3r1LfPykSZP0xhtvlGseCxYs0MyZM8t83MKFC/XRRx+Va9qX4mKvD6gI0dHR6tOnj/r1\n6yefz6cePXpo4MCBSk9PL/HxoX5Gzjd8+HAdP378cuNWGk+4AwAl2bp1q373u9/pD3/4gxo1aqRV\nq1bp2Wef1fz583XgwAFFRUVp5cqVFzxv6dKlWrNmjdauXau3335b48eP11/+8he1aNFCzZo1K3V+\nO3bs0Pjx45WUlKRbbrlFkvTBBx9o8ODB+vDDD9WgQYMKeV2LFy+WJGVlZVXI9Mpr69atuvHGG8My\nb6AiuVwuJScn68orrwze9vrrr2vmzJl65513Kmw+mzdvrrBpVQZKHI60e/du3X333WrUqJEkqXv3\n7poyZYr8fr/+9re/ye12a8iQITp+/Lji4+P1+OOPy+VyqUaNGjpz5oxycnKCP7/xxhtl7tkmJiZq\n1KhRwQKXpN69e6t27doKBALavn27Zs2apSuuuEJnzpzRH//4R23atEmLFi2S3+9X7dq1NX78eN12\n2206deqUpkyZoq+++koNGzZURESEYmNjJUldunRRYmKiJOn06dN64okndOjQIdWvX18zZsxQkyZN\niuXat2+fXnjhBR0/flyBQECDBw9W//79L/paJk2apDp16mjv3r367rvv1KxZM7300ktasWKF0tPT\n9eKLL8rtdqtjx476zW9+ox07digQCOiWW27RlClTVKdOHXXp0kW33nqr9u7dq1GjRunVV1/VBx98\nIEk6efKkunbtqvXr1ys1NVWLFy+W3+/X//3f/6lv374aM2ZMsTypqamaM2eOAoGAXC6XHn30UXXr\n1i2EtQAonTFG545VVlBQoCNHjigqKqrU5+zbt09DhgxRZmamrrnmGr300ku65pprtGHDhmLrcb9+\n/fTEE09o0qRJkqQhQ4Zo6dKluvbaayv9dZWbARxox44dpnPnzubIkSPGGGOSk5NNdHS0yczMNH/4\nwx/MzJkzTX5+vjl58qR58MEHze9//3tjjDFr1qwxPp/PPPLIIyYrK8u89NJLZtWqVWXO7/bbbzf/\n+Mc/Sr1/27ZtpmXLlubo0aPGGGMyMjJMr169zPHjx40xxnz99demXbt2Jjc318yaNctMnDjRGGNM\nVlaW6dixo0lMTDTGGNO5c2eTnp4enN4XX3xhjDHm3XffNQMHDjTGGDNx4kTz+uuvG7/fb3r27Gl2\n795tjDHm5MmT5t577zVffvnlBfkSExPNjBkzgs//5S9/afLz801+fr7x+XxmxYoVxhhjHn74YbN2\n7VpjjDELFiwwL774YnAa8+bNM9OmTQvmTEpKCt7XtWtXk56ebowxZvny5ebpp582xhgzZMgQc/Dg\nQWOMMceOHTMtW7Y02dnZZtu2baZXr17GGGOGDh1qVq9ebYwxZs+ePWb69OkXfzOAENx8882md+/e\npk+fPqZ9+/ama9euZubMmSYrK6vExycmJpq4uDiTnZ1tjDHm8ccfD67jpa3HRfMp+pw7EXvicCSv\n16tRo0Zp1KhRcrvduu+++3TllVeqRo0aGjhwYPBxdevW1X/+538qOTlZQ4YMUffu3dW9e3dJ0qFD\nh/Tll19qzJgxeuGFF3T48GHdeeedGjZs2AXzc7vdxb7Vl+RnP/uZfvazn0kqPMT2z3/+U8OGDQs+\nz+PxKCMjQ5999pkmT54sSbrqqqsUFxdX4vRuvvlm3XrrrZIkn8+nadOm6dSpU8H7MzIydOjQIT3z\nzDPBeZw9e1a7d+9WTEzMRbN26NBBHk/hx/umm24qdj1B0bQ+/vhjnTx5Mni40O/36+qrrw4+zuv1\nBn++7777tHLlSrVq1UorVqzQ+PHjJUmvvvqqPv74Y6WkpASvOcjNzS2WJSEhQdOnT9dHH32ktm3b\n6sknn7xodiBURYfT//d//1ePPPKIbr/9dl111VWlPr5t27bBPfXo6Ojgaa3S1uOix5a1bQgnShyO\ndPr0ad1555267777JBWeQ37llVd05ZVXatWqVYqOjtbNN98sqfADVqNGjQumMXv2bE2YMEFbtmxR\nTk6OFi5cqOHDh6tr165q3Lhxscfedttt+tvf/qYWLVoUu3369Onq1q2bIiIiFBkZGbw9EAjo7rvv\n1rx584K3fffdd2rUqJFcLlexD31RmZ7P7f7xulJjTPB0QJGCggLVr1+/2Ln/rKws1atXr/QF9y+1\na9cO/nx+nnOnP3nyZHXo0EFS4Ubr7NmzwfvPfb39+/dX//79NWDAAJ08eVJ33nmncnNz5fP51K1b\nN3m9Xg0YMEDr1q27YF4PPPCAunTpos2bN2vjxo1asGCBUlJSVLdu3TJfB3AxRevaLbfcokmTJmny\n5Mm67bbbdN1112nEiBH6/vvv5XK59MQTT0hSsc+Xy+WSVLje9+vXT927d7/oeuxUXJ0OR/r+++81\nePDg4J5pUlKSevXqJUn6+uuvlZiYqEAgoDNnzuitt97SvffeW+z5GzZs0LXXXqvo6Gjl5eUVK9Iz\nZ85cML+RI0cqKSlJu3fvDt62YsUKrV27Nvhl4Vxt2rTR5s2bg9/aP/nkE/Xt21d5eXnq0KGD/vSn\nP8kYoxMnTmj9+vUlvsY9e/Zoz549kqR3331Xd9xxh2rVqhW8/4YbblCtWrWUkpIiSTp69Kh69eql\nXbt2lb0AS+HxeOT3+yUV7q2//fbbys/PVyAQ0OTJk4t9KTnXtddeq3//93/Xc889FzwScvDgQZ0+\nfVpjx45Vp06dtG3bNuXn56ugoKDYcx988EHt3r1b/fr10/Tp03Xy5En98MMPl/wagJL07NlTd9xx\nh2bNmiVJWrJkid5//32tXLlSnTt3LvV5Bw8eVE5OTqnr8bmfGSdiTxyOdMMNN2jEiBG6//77ZYxR\nbGysnnvuOUnS6NGjNWPGDPXu3Vt+v18JCQkaMGBA8Ll5eXlatGiRli5dKklq166dli9frn79+unW\nW28t8epsr9ermTNnaubMmcrNzVV+fr4aN26sN998s8TDcy1atND06dM1btw4SVJERIReffVV1a5d\nW//zP/+jqVOnKiEhQVdffXWxLwFF3/4lqXnz5lq4cKEOHTqka665RnPmzCk2jxo1aigpKUkzZ87U\na6+9poKCAj355JO6/fbbL3m5du7cWXPmzFFeXp5GjRql2bNny+fzBS9smzBhwgU5i9x///0aM2aM\nFi1aJKnwdECnTp3Uo0cP1a9fX02aNFGLFi106NChYns8Tz/9tGbNmqVXXnlFLpdLo0eP1nXXXXfJ\nrwGQSl5Hp0yZor59+2rz5s1q165dSNOJjo5Wx44dS1yPGzdurLi4OA0aNEhJSUkXHKlzApex5ZgB\nAAAohsPpAABYihIHAMBSlDgAAJbiwrYqkpl5MuTHxsa2ltvt0o4df6/ERJevQYNIZWfnhDtGqZye\nTyJjRXF6xorI17Bh2X9aWJXK2qY59T1xYq6yMl3svWdP3IF27kxXRkZGuGOUyeOJCHeEi3J6PomM\nFcXpGZ2erzI49TU7MdflZKLEAQCwFCUOAIClKHEAACxFiQMAYClKHAAAS1HiDhQb21pNmzYNdwwA\ngMNR4gAAWIoSBwDAUpQ4AACWosQBALAUJQ4AgKUocQeyZex0AEB4UeIAAFiKEgcAwFKUOAAAlqLE\nAQCwFCUOAIClKHEHYux0AEAoKHEAACxFiQMAYClKHAAAS1HiAABYihIHAMBSlLgDMXY6ACAUlDgA\nAJaixAEAsBQlDgCApShxAAAsRYkDAGApStyBGDsdABAKShwAAEtR4gAAWIoSBwDAUpQ4AACWosSB\nMEtNdWv+/JpKTeXjCKB8POEOgAvt3Jmuhg3rKTPzZLijVIpBg67QunVVterVq6L5XI46//p/rbCm\nuLjKX45xcX4tX55b6fMBqhNK3CL33BOpPXsiwh3jPDaUJGywbp1HjRpdzvrkjHUxOrpAGzfmhDsG\nfiIuWuIFBQUaNmyY/H6/lixZonr1LvyQdOnSRX/5y180depU9ezZU+3bt6+0sBezd+9e/fDDD/J6\nvXrqqac0Z84ceTyX/x2ldevWuuuuu/S73/1OknTw4EGNHj1aH3zwgXJycvToo4/qwIED+vTTTy97\nXmVx2obB6UcLnJ5Pkvbtq6cOHYz8fpc8HqOUlBx5vYFwxyrGhuVoQ0agMly05Y4dO6acnBy99957\npT7G5XJVeKhLsXbtWl1zzTXyer367W9/W2HTjYqKChb4qlWr9Oabbyo7O1uSFBkZqeTk5LB9cYF9\nUlPd2rLFo7Zt/fJ6A2rTRkpJySl2G2C71FS30tKkmBg363Qlu2iJP//88zp48KCmTp2qX/3qV3r6\n6ad16tQpFRQUaOzYsbrrrrsueI7f79ekSZP0zTffyBijoUOHqlmzZnr55Ze1aNEirV69WosXL1ZK\nSoo+//xzvf/++2rUqJEaNmyoBx54QPv379fUqVOVnJysnj17KjY2Vv/4xz8UFRWlefPmqXbt2hfM\n89ixY1qxYoVq1qypli1bauzYscGjAx6PR0eOHFFeXp7uvfdebdiwQUePHlVSUpIaN26sefPmaefO\nncGjDj169Ch1eURFRentt99WXFzcJSxqVO258CLOOMR6oXPPf/+0z4lzLrx6SU11q0+fSPn9kscT\n6cijS9XJRbeoU6dO1VNPPaVp06Zpzpw5ateunQYPHqxjx47poYce0rp16y54zrvvvqurr75ac+fO\n1enTp9W/f3+98847OnLkiPLz87Vp0yZFREQoKytL69evV/fu3fXFF18Um0bR3n1ubq769u2r2NhY\nzZ07V++8846GDRt2wTyvvfZa9e/fXw0bNlRMTEyxowPXX3+9ZsyYoalTp+rw4cNasmSJEhMTtWHD\nBjVt2lTffvut3n77beXl5en+++9X+/btVbdu3RKXR8eOHctcoKVp0CBSHk/o57Nbt5Z27XJqAZ3L\nhoxwsss/F14kPOtiq1ZSenrZj2vYsHp9VkrbpqWlSX5/4c9+v0tpaXWUkFDF4crgxPfiUjOFvFu0\nf/9+9e3bV1JhadatW1dZWVkXPG7fvn1q27atJKlOnTpq3ry5vvnmG7Vv315bt27V0aNH1bt3b23Z\nskU7d+7UuHHjipW4MSb4c40aNRQbGytJuv3227Vp06Zyv8CWLVtKkurXr6/mzZsHfz579qz27t2r\nXbt2aciQITLGqKCgQN9++62io6PLPZ+yZGeHfj47Nra13G6Xvv/+7xWeoyI5/Tyk0/L9uIfy4/nv\nhIQ6jspYEqctx5KEO2Nm5sXvr4h8Tiue0rZpMTFueTw/rucxMTnKzHTOnni415WSlJXpYu99yH+Y\n2qxZM+3YsUNS4eHrH374QVFRUcVKV5KaN2+u1NRUSdKpU6f09ddf6/rrr1dcXJyWLFmi6OhotW/f\nXm+99ZaaNGmiiIgI1axZU5n/+hTs2rUrOK38/Hx99dVXkqTPP/9cN954Y6n5XC6XAoHCFeXcTBc7\nZ9+8eXPdddddevPNN/Xmm2+qR48e+rd/+7dQFwlQLl5vQCkpOZoy5SyHGFFtFa3ns2eL9bwKhLwn\nPnLkSD3zzDNas2aNzp49qxkzZigiIuKCkrz//vv17LPPatCgQTp79qxGjx6tq666Sg0aNFBGRoZG\njBihm266SUePHtWIESMkSffee6/Gjh2r7du3q1WrVsWmt3TpUh05ckTXXXednnzyyVLztW7dWnPn\nzlWzZs1KLO6SbuvcubO2bdumhx56SLm5uYqLi1NkZGSoiwQoN683IK83L9wxgErl9QaUkCBH7YFX\nVy5z/q60g3Tp0kVr1qxRjRo1wpahffv2Zf75WCiPKc/hm6LD6Tt2cDj9cjg9n0TGiuL0jNXxcHpZ\nr8ep74kTc13O4XRHD/bicrmKHRo/evSoxo8fH9yrNsbI5XLpP/7jPzR69OhKyXDixAkNHz48+Gdm\n5yr6O3Gn/JkdAOCnxdElvn79+mK///znP1dycnKVZvj730vfGy76O3EAAMKBf3HBgXbuTFdGRka4\nYwAAHI4SBwDAUpQ4AACWosQBALAUJQ4AgKUocQAALEWJO1BsbGs1bdo03DEAAA5HiQMAYClKHAAA\nS1HiAABYihIHAMBSlDgAAJaixB2IsdMBAKGgxAEAsBQlDgCApShxAAAsRYkDAGApShwAAEtR4g7E\n2OkAgFBQ4gAAWIoSBwDAUpQ4AACWosQBALAUJQ4AgKUocQdi7HQAQCgocQAALEWJAwBgKUocAABL\nUeIAAFiKEgcAwFKUuAMxdjoAIBSUOAAAlqLEAQCwFCUOAIClKHEAACxFiQMAYClK3IEYOx0AEApK\nHAAAS1HiAABYihIHAMBSlDgAAJaixAEAsBQl7kCMnQ4ACAUlDgCApShxAAAsRYkDAGApShwAAEtR\n4gAAWIoSdyDGTgcAhIISBwDAUpQ4AACWosQBALAUJQ4AgKUocQAALEWJOxBjpwMAQkGJAwBgKUoc\nAABLUeIAAFiKEgcAwFKUOAAAlqLEHYix0wEAoaDEAQCwFCUOAIClKHEAACxFiQMAYClKHAAAS1Hi\nDsTY6QCAUFDiAABYihIHAMBSlDgAAJaixAEAsBQlDlSx1FS35s+vqdRUPn4ALo8n3AFwoZ0709Ww\nYT1lZp4Md5RKMWjQFVq3rqpWvXpVNJ9LUetf/3dyxiIVkzEuzq/ly3MrZFoAKHHHat1a2rXrp7Nx\nx0/DunUeNWpUWetM1a6L0dEF2rgxp0rnCZyPEi/DypUrNX/+fA0YMECpqany+/2SpOnTp2vv3r16\n+eWX1a1bN40bN65C55ueLsfviTv9aIET86WmutWnT6T8fpc8HqNNm1xq3txZGc/nxOV4PhsyApWB\nEg9B79699e233+rhhx9W165d9emnn2revHlKTExUTk6ODhw4EO6IsITXG1BKSo62bPGobVu/2rSp\no8zMwnIvus3rDYQ7JnBZUlPdSkuTYmLcrM+VjBIP0YQJE1SvXuHhOr/fr1q1apXxDFyKij9f7uTD\n/SWdE3fqelWxy5Fz49XXj0ebJI8nUikpORR5JaLEQxQVFSVJ2r9/v+bOnaukpKRyPb9Bg0h5PBHl\nek7DhoUbzsLz4+V6ahVycknCqSrn3Hj418VWrQpPhZWk6PNcXZS2TUtLk/511lF+v0tpaXWUkFDF\n4crgxPfiUjNR4uWwdetWzZgxQ3PnzlWTJk3K9dzs7NAvgImNbS2326UdO/4uSdqwoVyzqjJOPw/p\n9HxSYcYPPzxd7Dy50/ZcbFmOTsmYmXnhbRWRz2nFU9o2LSbGLY/nx/U5JiZHmZmszxdTVqaLvfeU\neIi2bdumF154Qa+99pp+/vOfhzsOqpHzz5M7qcCB8ipan9PS6igmxllfSKsjSjwExhj9+te/lt/v\n14QJE2SMUbNmzTRt2rRwR0M14fUG5PXmhTsGUCG83oASEuSoPfDqihIP0fvvvx/uCAAAFMO4jyFY\nvXq1li1bdsHta9as0dKlS6s+EAAAYk+8TD6fTz6fr8T74uPjFR8fX8WJAAAoxJ64A+3cma6MjIxw\nxwAAOBwlDgCApShxAAAsRYkDAGApShwAAEtR4gAAWIoSd6DY2NZq2rRpuGMAAByOEgcAwFKUOAAA\nlqLEAQCwFCUOAIClKHEAACxFiTsQY6cDAEJBiQMAYClKHAAAS1HiAABYihIHAMBSlDgAAJaixB2I\nsdMBAKGgxAEAsBQlDgCApShxAAAsRYkDAGApShwAAEtR4g7E2OkAgFBQ4gAAWIoSBwDAUpQ4AACW\nosQBALAUJQ4AgKUocQdi7HQAQCgocQAALEWJAwBgKUocAABLUeIAAFiKEgcAwFKUuAMxdjoAIBSU\nOAAAlqLEAQCwFCUOAIClKHEAACxFiQMAYClK3IEYOx0AEApKHAAAS1HiAABYihIHAMBSlDgAAJai\nxAEAsBQl7kCMnQ4ACAUlDgCApShxAAAsRYkDAGApShwAAEtR4gAAWIoSdyDGTgcAhIISBwDAUpQ4\nAACWosQBALAUJQ4AgKUocQAALEWJOxBjpwMAQkGJAwBgKUocAABLUeIAAFiKEgcAwFKUOAAAlqLE\nHYix0wEAoaDEAQCwFCUOAIClKHEAACxFiQMAYClKHHCQ1FS35s+vqdRUPpoAyuYJdwBcaOfOdDVs\nWE+ZmSfDHSXsBg26QuvWXc5qWq/CslSekjLWqvIUF1f5yzEuzq/ly3MrfT5AdUKJVzP33BOpPXsi\nqnCONpQkbLBunUeNGl3O+uT0dfHHfNHRBdq4MSeMWVBdUOJlWLlypebPn68HHnhAW7duVX5+vqKi\nojR37lxt2rRJL7/8srp166Zx48aFO6okVemGwelHC5yeTyqeMTXVrT59IuX3u+TxGKWk5MjrDYQ5\noX3L0Ymcng/2osRD0Lt3b2VnZ8vn86lv375asGCB/vjHP2ro0KHKycnRgQMHwh0R1YDXG1BKSo62\nbPGobVu/IwocgLNR4iGaNGmSJCkQCOjo0aP6xS9+EeZEsFVqqjtY1AkJxe/zegPyevPCEwyoIKmp\nbqWlSTExbr6MVjJKvBz8fr/69u2rvLw8jR49OtxxrHf5F62FwsnnSYsuXnNyxiIVl5EL2Kq3H08L\nSR5PpGNOC1VXlHg5eDwerV69Wp999pnGjx+v5OTkkJ/boEGkPJ7QLjgrGjc9IyMjeFvr1tKuXeVJ\nW1VsKCA4yeVfwFaa8KyLrVpJ6ellP65hw+r1WSltm5aWJvn9hT/7/S6lpdW54IhTuDnxvbjUTJR4\nCIwxmj59uuLj43XXXXcpMjJSbnf5/o43Ozv0C84CASO321XsQpgNG8o1uyrh9It1nJjv/IvXNm1y\nqXlzZ2U8nxOX4/nCnTEz8+L3V0Q+pxVPadu0mBi3PJ4f1/GYmBxlZjpnTzzc60pJysp0sfeeEg+B\ny+XS4MGD9dxzzykpKUlut1tTp04NdyxY6PyL19q0qVNmAQA2KVrH09LqKCaGQ+mVjRIPgTFGN9xw\nQ7kOnwOl4eI1VHdeb0AJCXLUHnh1xdiOIVi9erWWLVt2we1r1qzR0qVLqz4QAABiT7xMPp9PPp+v\nxPvi4+MVHx9fxYkAACjEnrgD7dyZXuzKdAAASkKJAwBgKUocAABLUeIAAFiKEgcAwFKUOAAAlqLE\nHSg2tnVw/HQAAEpDiQMAYClKHAAAS1HiAABYihIHAMBSlDgAAJaixB2IsdMBAKGgxAEAsBQlDgCA\npShxAAAsRYkDAGApShwAAEtR4g7E2OkAgFBQ4gAAWIoSBwDAUpQ4AACWosQBALAUJQ4AgKUocQdi\n7HQAQCgocQAALEWJAwBgKUocAABLUeIAAFiKEgcAwFKUuAMxdjoAIBSUOAAAlqLEAQCwFCUOAICl\nKHEAACxFiQMAYClK3IEYOx0AEApKHAAAS1HiAABYihIHAMBSlDgAAJaixAEAsBQl7kCMnQ4ACAUl\nDgCApShxAAAsRYkDAGApShwAAEtR4gAAWIoSdyDGTgcAhIISBwDAUpQ4AACWosQBALAUJQ4AgKUo\ncQAALEWJOxBjpwMAQkGJAwBgKUocAABLUeIAAFiKEgcAwFKUOAAAlqLEHYix0wEAoaDEAQCwFCUO\nAIClKHEAACxFiQMAYClKHAAAS1HiDsTY6QCAUFDiAABYihIHAMBSlDgAAJaixAEAsBQlDgCApShx\nB2Ls9J+21FS35s+vqdRUPp4ALs4T7gBAWQYNukLr1l3qqlqvQrNUjtIy1qrSFBdXNcsxLs6v5ctz\nq2ReQHVAiZdh5cqVmj9/voYOHaphw4Zp+/btGj9+vD7++GOtWbNGL7/8srp166Zx48aFNec990Rq\nz56IMMzZhpKELdat86hRo0tdp5y+Lhbmi44u0MaNOWHOguqCEg9B7969NWzYMH333XdatmyZ/H6/\nJCk+Pl45OTk6cOBAmBMqLBuFhg3rKTPzZJXPN1ROzyddmDE11a0+fSLl97vk8RilpOTI6w2EMaGd\ny9FpnJ4P9uKkW4jy8vL0/PPP6/nnnw93FFRjXm9AL754Rl26+PXii2fCXuDApUhNdWvOHHFdRxVg\nTzwExhhNnz5d//Vf/6VGjRqFO85Pzk/1nPhHH3kU5rM05+CcOELz49EkyeOJdMTRpOqMEg/BiRMn\ntHPnTh06dEjGGB0/flxPPfWUfvvb34Y8jQYNIuXxhHbOumjc9FCvUG/dWtq1K+QoFcyGkoQtqvM5\n8Vat6ik9PdwpKk5p27S0NOlfZxzl97uUllZHCQlVHK4MDRs6b1251EyUeAiioqL04YcfBn9v3759\nuQpckrKzQz9nHQgYud2ukM+hbdhQrigVxunn+ZyeT+KceEVxesaifJmZlzcNJyltmxYT45bH8+M6\nHBOTo8xM5+yJO3FdKSvTxd57ShxwEK83oJSUHG3Z4lHbtv6wFzhQXkXrcFpaHcXEhP9LaHVHiYfA\nGFPs908//TRMSfBT4PUG5PXmhTsGcMm83oASEuSoPfDqiksHQ7B69WotW7bsgtvXrFmjpUuXVn0g\nAADEnniZfD6ffD5fiffFx8crPj6+ihMBAFCIPXEHYux0AEAoKHEAACxFiQMAYClKHAAAS1HiAABY\nihIHAMDMELTYAAAGo0lEQVRSlLgDxca2Do6fDgBAaShxAAAsRYkDAGApShwAAEtR4gAAWIoSBwDA\nUpS4AzF2OgAgFJQ4AACWosQBALAUJQ4AgKUocQAALEWJAwBgKUrcgRg7HQAQCkocAABLUeIAAFiK\nEgcAwFKUOAAAlqLEAQCwFCXuQIydDgAIBSUOAIClKHEAACxFiQMAYClKHAAAS1HiAABYihJ3IMZO\nBwCEghIHAMBSlDgAAJaixAEAsBQlDgCApShxAAAs5TLGmHCHAAAA5ceeOAAAlqLEAQCwFCUOAICl\nKHEAACxFiQMAYClKHAAAS1HiAABYyhPuACjOGKPnn39eX331lWrWrKlZs2apcePG4Y4lSfryyy/1\nm9/8RsnJyTp06JAmTpwot9utG2+8UVOnTg1rNr/fr2eeeUaHDx9Wfn6+Ro4cqRYtWjgqYyAQ0JQp\nU3TgwAG53W5NmzZNNWvWdFRGScrKytJ9992nN954QxEREY7L179/f9WtW1eSdP3112vkyJGOy7hk\nyRJ99NFHys/P16BBg3TnnXc6LmNlOndb4cQcy5Yt05/+9CddddVVkqTp06dX2b8cWdK2qkuXLpc+\nQQNHWbt2rZk4caIxxpgvvvjCPPbYY2FOVGjp0qWmV69e5oEHHjDGGDNy5EizY8cOY4wxzz33nPnr\nX/8aznjmvffeMy+88IIxxpgTJ06YTp06OS7jX//6V/PMM88YY4zZtm2beeyxxxyXMT8/34waNcrE\nx8eb/fv3Oy7f2bNnjc/nK3ab0zJu27bNjBw50hhjzOnTp01iYqLjMlam87cVTszxq1/9yuzatSsM\nqYpvq44fP246dep0WdPjcLrD7Ny5Ux06dJAk3XrrrUpPTw9zokJNmjTRwoULg7/v2rVLXq9XknTP\nPffos88+C1c0SVJCQoLGjBkjSSooKFBERIR2797tqIxxcXGaMWOGJOnIkSO68sorHZdxzpw5+uUv\nf6lGjRrJGOO4fHv27FFOTo6GDx+uYcOG6csvv3Rcxk8//VQ33XSTHn/8cT322GPq1KmT4zJWpvO3\nFU7MsWvXLi1evFiDBg3SkiVLqjTXuduqQCAgj+fHA+JdunRRXl5euabH4XSHOXXqlOrVqxf83ePx\nKBAIyO0O7/etbt266fDhw8HfzTmj9dapU0cnT54MR6ygK664QlLh8hszZoyefPJJzZkzJ3i/EzJK\nktvt1sSJE7Vu3Tq98sor2rx5c/C+cGdcsWKFrr76arVr106LFi2SVLiRKRLufJJUu3ZtDR8+XAMH\nDlRGRoYeeeQRx62L2dnZOnLkiBYvXqxvvvlGjz32mOOWY2U6f1vhxBw9e/bUQw89pLp162rUqFH6\n5JNP1LFjxyrJdf62auzYsZo0aZK+/fZbZWVlafjw4fJ4PHrjjTdCmh4l7jB169bV6dOng787ocBL\ncm6m06dPq379+mFMU+jo0aMaPXq0Hn74YfXs2VNz584N3ueUjJI0e/ZsZWVlacCAATp79mzw9nBn\nXLFihVwulzZv3qyvvvpKEyZMUHZ2tmPySVLTpk3VpEmT4M9RUVHavXt38H4nZIyKilLz5s3l8Xh0\nww03qFatWjp27Fjwfidk/KkbOnRo8LqKjh07avfu3VVW4tKF26qePXtKKtwTf/3111WjRo2Qp+W8\ndviJu+OOO/TJJ59Ikr744gvddNNNYU5UspYtW2rHjh2SpI0bNyo2Njasef75z39q+PDhevrpp+Xz\n+SRJt9xyi6Myrlq1KnjorlatWnK73WrdurW2b98uKfwZ33rrLSUnJys5OVnR0dF68cUX1aFDB0ct\nw/fee0+zZ8+WJB07dkynTp1Su3btHLMMJSk2NlabNm2SVJgxNzdXbdq0cVTGqmAc8m9rnZ/j1KlT\n6tWrl3Jzc2WM0datW9WqVasqy1PStqqIy+Uq93JjT9xhunXrps2bN+vBBx+UJP36178Oc6KSTZgw\nQc8++6zy8/PVvHlz9ejRI6x5Fi9erB9++EFJSUlauHChXC6XJk+erJkzZzomY/fu3TVp0iQ9/PDD\n8vv9mjJlipo1a6YpU6Y4JuP5nPY+DxgwQJMmTdKgQYPkdrs1e/ZsRUVFOWoZdurUSampqRowYEDw\nr01+8YtfOCpjVXC5XOGOIOnHHH/+85+Vm5urgQMHaty4cRo8eLBq1aqlu+++W/fcc0+V5SlpW/Xa\na6+pZs2aWr9+fbmnxz9FCgCApTicDgCApShxAAAsRYkDAGApShwAAEtR4gAAWIoSBwDAUpQ4AACW\n+n8kjpZZCUTCfAAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x10b14fda0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.Matplot.summary_plot(uae_model.followup_time)"
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {
"collapsed": false,
"scrolled": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAADA0v8HY9LLq1avxqYAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x110634a20>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.Matplot.summary_plot(uae_model.μ, custom_labels=plot_labels)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Follow-up time effect size estimates. Positive values indicate higher probability of event with increased follow-up time."
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYAAAAJYIDAAAwBKBAQAAWCIwAAAAS/8PZpL8tPRBjlYAAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x1109244e0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.Matplot.summary_plot(uae_model.β_fup, custom_labels=plot_labels)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Age effect size estimates. Positive values suggest higher probability of event with each year above age 40."
]
},
{
"cell_type": "code",
"execution_count": 55,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.gridspec.GridSpec at 0x12584a4e0>"
]
},
"execution_count": 55,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAASUVORK5CYII=\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x12584a518>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.forestplot(trace_uae, vars=['β_age'], ylabels=plot_labels)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Estimated probabilities of follow-up interventions for 6-month followup and age 40."
]
},
{
"cell_type": "code",
"execution_count": 57,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.gridspec.GridSpec at 0x128ada4a8>"
]
},
"execution_count": 57,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x128ada828>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.forestplot(trace_uae, vars=['p_6'], ylabels=plot_labels)"
]
},
{
"cell_type": "code",
"execution_count": 58,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"p_6:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.023 0.007 0.000 [0.012, 0.034]\n",
" 0.033 0.007 0.000 [0.020, 0.048]\n",
" 0.016 0.004 0.000 [0.009, 0.025]\n",
" 0.000 0.002 0.000 [0.000, 0.001]\n",
" 0.000 0.001 0.000 [0.000, 0.001]\n",
" 0.000 0.004 0.000 [0.000, 0.001]\n",
" 0.927 0.013 0.001 [0.907, 0.948]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.013 0.019 0.022 0.026 0.036\n",
" 0.020 0.027 0.032 0.037 0.048\n",
" 0.009 0.013 0.016 0.019 0.026\n",
" 0.000 0.000 0.000 0.000 0.002\n",
" 0.000 0.000 0.000 0.000 0.002\n",
" 0.000 0.000 0.000 0.000 0.002\n",
" 0.903 0.921 0.928 0.934 0.945\n",
"\n"
]
}
],
"source": [
"pm.summary(trace_uae, vars=['p_6'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Estimated probabilities of follow-up interventions for 12-month followup and age 40."
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.gridspec.GridSpec at 0x11eb50a58>"
]
},
"execution_count": 59,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x11a779be0>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.forestplot(trace_uae, vars=['p_12'], ylabels=plot_labels)"
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"p_12:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.029 0.008 0.000 [0.015, 0.042]\n",
" 0.032 0.006 0.000 [0.020, 0.044]\n",
" 0.019 0.005 0.000 [0.010, 0.028]\n",
" 0.000 0.002 0.000 [0.000, 0.001]\n",
" 0.000 0.001 0.000 [0.000, 0.001]\n",
" 0.000 0.004 0.000 [0.000, 0.001]\n",
" 0.920 0.013 0.001 [0.899, 0.940]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.016 0.024 0.028 0.033 0.044\n",
" 0.021 0.027 0.031 0.036 0.045\n",
" 0.011 0.016 0.019 0.022 0.029\n",
" 0.000 0.000 0.000 0.000 0.002\n",
" 0.000 0.000 0.000 0.000 0.002\n",
" 0.000 0.000 0.000 0.000 0.001\n",
" 0.897 0.914 0.921 0.927 0.938\n",
"\n"
]
}
],
"source": [
"pm.summary(trace_uae, vars=['p_12'])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Estimated probabilities of follow-up interventions for 12-month followup and age 50."
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.gridspec.GridSpec at 0x1286a26a0>"
]
},
"execution_count": 61,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nblEhBQOAQgAAAABJRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x1286bb2e8>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.forestplot(trace_uae, vars=['p_6_50'], ylabels=plot_labels)"
]
},
{
"cell_type": "code",
"execution_count": 62,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"p_6_50:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.089 0.043 0.003 [0.007, 0.168]\n",
" 0.001 0.002 0.000 [0.000, 0.002]\n",
" 0.012 0.007 0.000 [0.001, 0.026]\n",
" 0.001 0.003 0.000 [0.000, 0.005]\n",
" 0.025 0.048 0.004 [0.000, 0.111]\n",
" 0.173 0.255 0.024 [0.000, 0.788]\n",
" 0.699 0.219 0.020 [0.170, 0.917]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.014 0.061 0.085 0.112 0.185\n",
" 0.000 0.001 0.001 0.001 0.003\n",
" 0.002 0.007 0.011 0.016 0.030\n",
" 0.000 0.000 0.000 0.000 0.008\n",
" 0.000 0.001 0.007 0.026 0.156\n",
" 0.000 0.011 0.040 0.211 0.849\n",
" 0.131 0.637 0.791 0.847 0.902\n",
"\n"
]
}
],
"source": [
"pm.summary(trace_uae, vars=['p_6_50'])"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Applied log-transform to nu and added transformed nu_log to model.\n",
"Applied interval-transform to σ and added transformed σ_interval to model.\n"
]
}
],
"source": [
"med_manage_model = specify_model(med_manage_model, 'med_manage')"
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: ϵ\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:571: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: <DisconnectedType>\n",
" handle_disconnected(rval[i])\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: ϵ\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:571: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: <DisconnectedType>\n",
" handle_disconnected(rval[i])\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: σ_interval\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:571: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: <DisconnectedType>\n",
" handle_disconnected(rval[i])\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: σ_interval\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:571: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: <DisconnectedType>\n",
" handle_disconnected(rval[i])\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: β_age\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: β_age\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:571: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: <DisconnectedType>\n",
" handle_disconnected(rval[i])\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: β_fup\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: μ\n",
" handle_disconnected(elem)\n",
"/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gradient.py:545: UserWarning: grad method was asked to compute the gradient with respect to a variable that is not part of the computational graph of the cost, or is used only by a non-differentiable operator: age_centered_missing\n",
" handle_disconnected(elem)\n"
]
},
{
"name": "stdout",
"output_type": "stream",
"text": [
"Assigned NUTS to followup_time_missing\n",
"Assigned NUTS to nu_log\n",
"Assigned NUTS to age_centered_missing\n",
"Assigned NUTS to μ\n",
"Assigned NUTS to β_fup\n",
"Assigned NUTS to β_age\n",
"Assigned NUTS to σ_interval\n",
"Assigned NUTS to ϵ\n",
"Assigned NUTS to followup_time_missing\n",
"Assigned NUTS to nu_log\n",
"Assigned NUTS to age_centered_missing\n",
"Assigned NUTS to μ\n",
"Assigned NUTS to β_fup\n",
"Assigned NUTS to β_age\n",
"Assigned NUTS to σ_interval\n",
"Assigned NUTS to ϵ\n"
]
},
{
"ename": "IndexError",
"evalue": "index out of bounds",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-46-5f5abf71fe8b>\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[0mmed_manage_model\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 2\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 3\u001b[0;31m \u001b[0mtrace_med_manage\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpm\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[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/sampling.py\u001b[0m in \u001b[0;36msample\u001b[0;34m(draws, step, start, trace, chain, njobs, tune, progressbar, model, random_seed)\u001b[0m\n\u001b[1;32m 122\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[1;32m 123\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 124\u001b[0;31m \u001b[0mstep\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0massign_step_methods\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstep\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 125\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 126\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mnjobs\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mNone\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/sampling.py\u001b[0m in \u001b[0;36massign_step_methods\u001b[0;34m(model, step, methods)\u001b[0m\n\u001b[1;32m 67\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 68\u001b[0m \u001b[0;31m# Instantiate all selected step methods\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 69\u001b[0;31m \u001b[0msteps\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvars\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mselected_steps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0ms\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mselected_steps\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mselected_steps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0ms\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 70\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 71\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msteps\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m==\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/sampling.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 67\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 68\u001b[0m \u001b[0;31m# Instantiate all selected step methods\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 69\u001b[0;31m \u001b[0msteps\u001b[0m \u001b[0;34m+=\u001b[0m \u001b[0;34m[\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mvars\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mselected_steps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0ms\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0ms\u001b[0m \u001b[0;32min\u001b[0m \u001b[0mselected_steps\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mselected_steps\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0ms\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 70\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 71\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msteps\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m==\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/step_methods/nuts.py\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, profile, **kwargs)\u001b[0m\n\u001b[1;32m 67\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 68\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---> 69\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 70\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 71\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/tuning/scaling.py\u001b[0m in \u001b[0;36mguess_scaling\u001b[0;34m(point, vars, model)\u001b[0m\n\u001b[1;32m 107\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[1;32m 108\u001b[0m \u001b[0;32mtry\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 109\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 110\u001b[0m \u001b[0;32mexcept\u001b[0m \u001b[0mNotImplementedError\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 111\u001b[0m \u001b[0mh\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mfixed_hessian\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[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/tuning/scaling.py\u001b[0m in \u001b[0;36mfind_hessian_diag\u001b[0;34m(point, vars, model)\u001b[0m\n\u001b[1;32m 101\u001b[0m \"\"\"\n\u001b[1;32m 102\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--> 103\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 104\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 105\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/memoize.py\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/Repositories/pymc3/pymc3/theanof.py\u001b[0m in \u001b[0;36mhessian_diag\u001b[0;34m(f, vars)\u001b[0m\n\u001b[1;32m 101\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 102\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 103\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 104\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 105\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mempty_gradient\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/theanof.py\u001b[0m in \u001b[0;36m<listcomp>\u001b[0;34m(.0)\u001b[0m\n\u001b[1;32m 101\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 102\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 103\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 104\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 105\u001b[0m \u001b[0;32mreturn\u001b[0m \u001b[0mempty_gradient\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/theanof.py\u001b[0m in \u001b[0;36mhessian_diag1\u001b[0;34m(f, v)\u001b[0m\n\u001b[1;32m 92\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 93\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 94\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 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/anaconda3/lib/python3.5/site-packages/theano/scan_module/scan_views.py\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 67\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 68\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---> 69\u001b[0;31m name=name)\n\u001b[0m\u001b[1;32m 70\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 71\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/scan_module/scan.py\u001b[0m in \u001b[0;36mscan\u001b[0;34m(fn, sequences, outputs_info, non_sequences, n_steps, truncate_gradient, go_backwards, mode, name, profile, allow_gc, strict)\u001b[0m\n\u001b[1;32m 473\u001b[0m \u001b[0;31m# If not we need to use copies, that will be replaced at\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 474\u001b[0m \u001b[0;31m# each frame by the corresponding slice\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 475\u001b[0;31m \u001b[0mactual_slice\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mseq\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'input'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mk\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mmintap\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 476\u001b[0m \u001b[0m_seq_val\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mtensor\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mas_tensor_variable\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mseq\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'input'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 477\u001b[0m \u001b[0m_seq_val_slice\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0m_seq_val\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mk\u001b[0m \u001b[0;34m-\u001b[0m \u001b[0mmintap\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/tensor/var.py\u001b[0m in \u001b[0;36m__getitem__\u001b[0;34m(self, args)\u001b[0m\n\u001b[1;32m 529\u001b[0m self, *theano.tensor.subtensor.Subtensor.collapse(\n\u001b[1;32m 530\u001b[0m \u001b[0margs\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 531\u001b[0;31m lambda entry: isinstance(entry, Variable)))\n\u001b[0m\u001b[1;32m 532\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 533\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mtake\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindices\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0maxis\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;32mNone\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mmode\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'raise'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gof/op.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self, *inputs, **kwargs)\u001b[0m\n\u001b[1;32m 649\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[1;32m 650\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 651\u001b[0;31m \u001b[0mrequired\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mthunk\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 652\u001b[0m \u001b[0;32massert\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mrequired\u001b[0m \u001b[0;31m# We provided all inputs\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 653\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gof/op.py\u001b[0m in \u001b[0;36mrval\u001b[0;34m()\u001b[0m\n\u001b[1;32m 852\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 853\u001b[0m \u001b[0;32mdef\u001b[0m \u001b[0mrval\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[0;32m--> 854\u001b[0;31m \u001b[0mfill_storage\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 855\u001b[0m \u001b[0;32mfor\u001b[0m \u001b[0mo\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[1;32m 856\u001b[0m \u001b[0mcompute_map\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mo\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[0;32mTrue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/anaconda3/lib/python3.5/site-packages/theano/gof/cc.py\u001b[0m in \u001b[0;36m__call__\u001b[0;34m(self)\u001b[0m\n\u001b[1;32m 1712\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0merror_storage\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfile\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0msys\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstderr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1713\u001b[0m \u001b[0;32mraise\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1714\u001b[0;31m \u001b[0mreraise\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mexc_type\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexc_value\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexc_trace\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 1715\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 1716\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;32m/Users/fonnescj/anaconda3/lib/python3.5/site-packages/six.py\u001b[0m in \u001b[0;36mreraise\u001b[0;34m(tp, value, tb)\u001b[0m\n\u001b[1;32m 684\u001b[0m \u001b[0;32mif\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__traceback__\u001b[0m \u001b[0;32mis\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mtb\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 685\u001b[0m \u001b[0;32mraise\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mwith_traceback\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtb\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 686\u001b[0;31m \u001b[0;32mraise\u001b[0m \u001b[0mvalue\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 687\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 688\u001b[0m \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mIndexError\u001b[0m: index out of bounds"
]
}
],
"source": [
"with med_manage_model:\n",
" \n",
" trace_med_manage = pm.sample(2000)"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"[followup_time_missing,\n",
" nu_log,\n",
" age_centered_missing,\n",
" μ,\n",
" β_fup,\n",
" β_age,\n",
" σ_interval,\n",
" ϵ]"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"med_manage_model.vars"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {
"collapsed": false
},
"outputs": [
{
"ename": "IndexError",
"evalue": "index 0 is out of bounds for axis 0 with size 0",
"output_type": "error",
"traceback": [
"\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[0;31mIndexError\u001b[0m Traceback (most recent call last)",
"\u001b[0;32m<ipython-input-44-ed6e9ecfa6a7>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mpm\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mforestplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mtrace_med_manage\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mvars\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'age_centered_missing'\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mylabels\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mplot_labels\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
"\u001b[0;32m/Users/fonnescj/Repositories/pymc3/pymc3/plots.py\u001b[0m in \u001b[0;36mforestplot\u001b[0;34m(trace_obj, vars, alpha, quartiles, rhat, main, xtitle, xrange, ylabels, chain_spacing, vline, gs)\u001b[0m\n\u001b[1;32m 378\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 379\u001b[0m \u001b[0;31m# Substitute HPD interval for quantile\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 380\u001b[0;31m \u001b[0mquants\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mvar_hpd\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mT\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m 381\u001b[0m \u001b[0mquants\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mvar_hpd\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mT\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m 382\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n",
"\u001b[0;31mIndexError\u001b[0m: index 0 is out of bounds for axis 0 with size 0"
]
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAegAAAFVCAYAAAAkBHynAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x1276cc518>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.forestplot(trace_med_manage, vars=['age_centered_missing'], ylabels=plot_labels)"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {
"collapsed": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"\n",
"p_6:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.552 0.000 0.000 [0.552, 0.552]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.552 0.552 0.552 0.552 0.552\n",
"\n",
"\n",
"p_12:\n",
"\n",
" Mean SD MC Error 95% HPD interval\n",
" -------------------------------------------------------------------\n",
" \n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.075 0.000 0.000 [0.075, 0.075]\n",
" 0.552 0.000 0.000 [0.552, 0.552]\n",
"\n",
" Posterior quantiles:\n",
" 2.5 25 50 75 97.5\n",
" |--------------|==============|==============|--------------|\n",
" \n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.075 0.075 0.075 0.075 0.075\n",
" 0.552 0.552 0.552 0.552 0.552\n",
"\n"
]
}
],
"source": [
"pm.summary(trace_med_manage, vars=['p_6', 'p_12'])"
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"<matplotlib.gridspec.GridSpec at 0x116f27c18>"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "iVBORw0KGgoAAAANSUhEUgAAAgwAAAF7CAYAAAC+SdPrAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3X98zfX///H7OTsz9ktkpC8tv0JqiflUeisR2hrlZ0Nb\nSck7KS2/kuoT423vlOTn8haS8nOrUflISuUT2+TdeUt+G6JYjMbMdpzn9w/sY/nxErPzwu16uXRp\nzo/XeZzz7LTbXq9jL4cxxggAAOAcnL4eAAAA2B/BAAAALBEMAADAEsEAAAAsEQwAAMASwQAAACwR\nDMBlZObMmXrggQfUvn17vfjiizp48GDRdXfeeafat29f9M+iRYskSbNnz9aDDz6o2NhY7dq1q+j2\nvXr10tatW8/5eD/88IOefPJJtW/fXm3btlXv3r21adOmi34ew4cP1/jx4yVJTz/9tLZs2aL09HS1\nbdv2jLd/6aWXNG3atL/0GOPHj1diYqLl7SZMmKBly5b9pW1fiHM9P+By4PL1AADOz8qVKzV16lTN\nnTtXlStX1ieffKJXXnlF77zzjrZt26ZrrrlGqampp91vypQp+p//+R8tWbJEs2bN0sCBA7V48WLV\nrl1bNWvWPOvjZWRkaODAgZo4caLq168vSVq4cKHi4uL0+eefq0KFCiXyvJKTkyVJ+/btK5Ht/VUr\nV65UnTp1fPLYwOWEYAAuE+vWrdNdd92lypUrS5Jat26toUOHyuPxaM2aNXI6nYqPj9eBAwfUpk0b\nPfPMM3I4HPL391d+fr7y8vKKvp42bZrlT+zjxo1Tnz59imJBktq2bauyZcvK6/UqPT1dI0aMULly\n5ZSfn6958+bp22+/1eTJk+XxeFS2bFkNHDhQDRs21KFDhzR06FBt2LBBYWFh8vPzU+PGjSVJLVq0\n0Lhx4yRJhw8f1nPPPacdO3YoNDRUw4cPV3h4eLG5tmzZopEjR+rAgQPyer2Ki4tThw4dzvlcXnrp\nJQUFBWnjxo367bffVLNmTY0ZM0YpKSlau3at/vnPf8rpdOree+/V6NGjlZGRIa/Xq/r162vo0KEK\nCgpSixYtdNttt2njxo3q06ePJk2apIULF0qScnNz1bJlS3355ZfKzMxUcnKyPB6P9u/fr4ceekjP\nP/98sXkyMzOVlJQkr9crh8Ohp59+Wq1atTqP/woA3+GQBHCZiIiI0KpVq/Trr79KkhYsWCCPx6MD\nBw7o2LFjuvvuu/Xee+/pww8/1HfffaeZM2dKkhISEhQfH68lS5boscce0+TJk9W9e3cFBgae8/HW\nrl2r22+//bTLW7VqpWuvvVaStHnzZr399tv6+OOPtXv3bo0ZM0ZTpkxRSkqKhg0bpmeffVb5+fl6\n5513VK5cOX3++ed6++23tW3btjM+5p49e9SzZ099/PHHiomJ0YABA4pdf+zYMT3//PPq37+/FixY\noJkzZ2rq1Klyu92Wr9+6dev03nvv6bPPPtPevXu1ePFide/eXbfccosGDRqk+++/X++++65cLpdS\nUlL08ccfq3LlynrzzTeLtnHTTTfp008/VXR0tI4cOaKffvpJkrRo0SI1b95cISEhmj59uv75z39q\n/vz5mj17tt59910dOHCg2Czjx49Xjx49tGDBAo0YMUIrV660nB/wNfYwAJeJyMhI9enTR3369JHT\n6VTHjh1Vvnx5+fv7q3PnzkW3Cw4OVo8ePTRz5kzFx8erdevWat26tSRpx44d+vHHH/X8889r5MiR\n2rVrl5o0aaLHH3/8tMdzOp2y+s3x1113na677jpJ0ooVK/T777/r8ccfL7qfy+VSVlaWvv/+e738\n8suSpIoVK+r+++8/4/bq1q2r2267TZLUvn17vf766zp06FDR9VlZWdqxY4eGDBlS9BhHjx7VunXr\nFBERcc5ZmzVrJpfr+P/ybrrppmKf/zi5ra+//lq5ublasWKFJMnj8RTFkXR8DU7q2LGjUlNT1aBB\nA6WkpGjgwIGSpEmTJunrr79WWlpa0WdEjhw5UmyWqKgoDRs2TMuWLVPTpk31wgsvnHN2wA4IBuAy\ncfjwYTVp0kQdO3aUdPyY/9ixY1W+fHl98sknqlevnurWrSvp+DdAf3//07YxatQoDRo0SP/7v/+r\nvLw8TZgwQT179lTLli1VvXr1Yrdt2LCh1qxZo9q1axe7fNiwYWrVqpX8/PyK7aXwer2666679NZb\nbxVd9ttvv6ly5cpyOBzF4uPkN+4/czr/b6enMabokMpJx44dU2hoaLHPauzbt08hISFnf+FOKFu2\nbNHXf57n1O2//PLLatasmaTj3+iPHj1adP2pz7dDhw7q0KGDOnXqpNzcXDVp0kRHjhxR+/bt1apV\nK0VGRqpTp05aunTpaY/1yCOPqEWLFlqxYoW++eYbjR8/XmlpaQoODrZ8HoCvcEgCuEzs3btXcXFx\nRT9xT5w4UTExMZKkTZs2ady4cfJ6vcrPz9cHH3yg6OjoYvf/6quvVKVKFdWrV08FBQXFvmnn5+ef\n9ni9e/fWxIkTtW7duqLLUlJStGTJkqIwOdWdd96pFStWFP1UvXz5cj300EMqKChQs2bNNH/+fBlj\ndPDgQX355ZdnfI7r16/X+vXrJUlz5sxRo0aNFBAQUHR9jRo1FBAQoLS0NEnSr7/+qpiYmKJDAxfC\n5XLJ4/FIOr4XYtasWSosLJTX69XLL79cLIBOVaVKFd1666169dVXi/bwbN++XYcPH1a/fv3UvHlz\nrVq1SoWFhTp27Fix+8bGxmrdunV6+OGHNWzYMOXm5uqPP/644OcAlAb2MACXiRo1aqhXr17q0qWL\njDFq3LixXn31VUnSs88+q+HDh6tt27byeDyKiopSp06diu5bUFCgyZMna8qUKZKku+++Wx9++KEe\nfvhh3XbbbWf8WwKRkZFKTExUYmKijhw5osLCQlWvXl3vv/++KlaseNrta9eurWHDhikhIUGS5Ofn\np0mTJqls2bLq27evXnvtNUVFRenaa68tFhwOh6Po61q1amnChAnasWOHKlWqpKSkpGKP4e/vr4kT\nJyoxMVH/+te/dOzYMb3wwgtn/KzF+brvvvuUlJSkgoIC9enTR6NGjVL79u2LPvQ4aNCg0+Y8qUuX\nLnr++ec1efJkSccPqTRv3lwPPPCAQkNDFR4ertq1a2vHjh3F9pQMGDBAI0aM0NixY+VwOPTss8/q\n+uuvv+DnAJQGB6e3BgAAVjgkAQAALBEMAADAEsEAAAAs8aHH85SdnVui26tQIVA5OXkluk1cHNbE\nXho3vkVOp0MZGf/x9Sg4Be8Teynp9QgLO/tfUWYPg4+4XH6+HgF/wprYy+rVa5WVleXrMfAnvE/s\npTTXg2AAAACWCAYAAGCJYAAAAJYIBgAAYIlgAAAAlggGALbUuPEtuvHGG309BoATCAYAAGCJYAAA\nAJYIBgAAYIlgAAAAlggGAABgiWAAYEucSwKwF4IBAABYIhgAAIAlggEAAFgiGAAAgCWCAQAAWCIY\nANgS55IA7IVgAAAAliyDITU1VW+99dYFbXzjxo3KzMy8oPuey8GDB7Vo0aIS3y4AADizS7qHYcmS\nJdq8eXOJb3f9+vVatmxZiW+3tGRmOpWUdPzfAABcDlznc6M1a9aoZ8+eysnJUWxsrObNm6d58+ZJ\nkl544QU98cQTWrp0qVatWiWv16vWrVurbdu2SklJUZkyZdSgQQPl5+drzJgx8vPz0w033KDXX39d\nCxcu1IIFC2SMUd++fXXgwAFNnz5dfn5+aty4sRISErR//34NHjxYf/zxhyQpKSlJycnJ2rBhg+bN\nm6emTZtqyJAh8nq9kqShQ4eqbt26at26tRo1aqSsrCzdcccdOnTokNxut2rWrKlRo0apTZs2mj9/\nvkJDQ/XRRx8pLy9PPXv2vEQv8//JzHSqXbtAeTySyxWotLQ8RUZ6L/njAgBwUYyFlJQU88QTTxhj\njPnll19MdHS0efTRR83mzZvNgQMHTOfOnY0xxrRo0cLs2rXLHD161MyZM8cYY8y4cePM7NmzjTHG\ntG7d2uzbt88YY8zbb79t5s6da1JSUswzzzxjjDHmwIEDJjo62uTn5xtjjBkwYIBZsWKFSUxMLNrG\nmjVrzMKFC016erpJSEgwxhjTt29fs2zZMmOMMT///LPp0KGDMcaYm2++2fz222+msLDQ3H777WbL\nli3GGGNatmxpcnNzzbhx48yHH35ojDEmNja2aLazKSz0WL1U52XUKGOk//tn1KgS2SxwxQkPDzfh\n4eG+HgPACee1h+Hmm2+WJIWFhenIkSPq0qWLUlJSdP3116tdu3aSpDfeeEOjR4/W77//rnvuuafY\n/ffv36/s7Gz169dPxhgVFBSoadOmuuGGG1SjRg1J0vbt27V//3499dRTMsYoLy9PO3fuVFZWljp1\n6iRJatiwoRo2bKj09PSibW/dulWRkZGSpHr16mnPnj2SpGuuuUZVqlSRJAUGBqpmzZqSpJCQEB09\nelQdOnRQQkKCIiMjFRYWpooVK57zNcjJyTufl8pSRIRTLlegPB6HXC6jiIg8ZWezh8EOwsJClJ2d\n6+sxcEJGxn9YExtiTeylpNcjLCzkrNedVzA4HI5iX7dp00ZTp05VhQoVNHbsWBUUFGjx4sVFH46M\njo7Wgw8+KIfDIa/XqwoVKqhq1aqaOHGigoODtWzZMgUFBWn37t1yOo8fx69WrZqqVq2qadOmyc/P\nT6mpqapfv762bdsmt9utunXrKiMjQ8uXL1fz5s117NgxSVKtWrWUkZGhFi1a6Oeff1alSpVOm/lU\nxhhJ0vXXX6+QkBBNnjxZHTt2PJ+XoURERnqVlpYntztIEREcjgAAXB7OKxj+rEyZMoqMjFROTo5C\nQ0MlSeXLl1eXLl1UtmxZNWvWTFWrVtUtt9yiN954Q7Vq1dLLL7+sXr16yev1KiQkRElJSdq9e3fR\nNitWrKgePXqoe/fu8nq9qlatmqKjo9WrVy8NGTJEaWlpcjqdGjFihPz9/bVp0ya9//77GjRokIYO\nHar33ntPHo9HI0eOPOfsp4ZEly5dNGLECI0ePfpCXoYLFhnpVVSU2LMAALhsOMzJH7n/omHDhqlN\nmza64447SnqmUrN48WJt2rRJffv2tbxtSe+CY7ee/bAm9sOa2A9rYi+2OyTxZz179lSFChUu61gY\nM2aMVq1apeTkZF+PAgCA7V1QMEydOrWk5yh1L7zwgq9HAADgssFvDgJgS5xLArAXggEAAFgiGAAA\ngCWCAQAAWCIYAACAJYIBAABYIhgA2NLq1WuVlZXl6zEAnEAwAAAASwQDAACwRDAAAABLBAMAALBE\nMAAAAEsEAwBb4lwSgL0QDAAAwBLBAAAALBEMAADAEsEAAAAsEQwAAMASwQDAljiXBGAvBAMAALBE\nMAAAAEsEAwAAsEQwAAAASwQDAACwRDAAsCXOJQHYC8EAAAAsEQwAAMASwQAAACwRDAAAwBLBAAAA\nLBEMAGyJc0kA9kIwAAAASwQDAACwRDAAAABLBAMAALBEMAAAAEsEAwBb4lwSgL0QDAAAwNJlGQwH\nDx7UokWLfD0GAABXjcsyGNavX69ly5b5eowLlpnpVFLS8X8DAHA5cFndIDU1VV999ZXy8/P1+++/\nKy4uTl9++aU2bdqkgQMHKi0tTWPHjpUkde3aVe+8846+//57vf/++woICFB4eLiGDRumhQsXnnU7\ngwYNUosWLfT5559rxowZ8vPzU+PGjZWQkKD9+/dr8ODB+uOPPyRJSUlJSk5O1oYNGzRv3jw1bdpU\nQ4YMkdfrlSQNHTpUdevWVevWrdWoUSNlZWXpjjvu0KFDh+R2u1WzZk2NGjVKbdq00fz58xUaGqqP\nPvpIeXl56tmz5yV8qY/LzHSqXbtAeTySyxWotLQ8RUZ6L/njAgBwUYyFlJQU88QTTxhjjPn0009N\nly5djDHGrFq1yvz97383UVFR5o8//jCbNm0yzzzzjMnJyTGtWrUyeXl5xhhj/vGPf5gPPvjgrNtZ\nuXKl6dOnjzlw4ICJjo42+fn5xhhjBgwYYFasWGESExPN7NmzjTHGrFmzxixcuNCkp6ebhIQEY4wx\nffv2NcuWLTPGGPPzzz+bDh06GGOMufnmm81vv/1mCgsLze233262bNlijDGmZcuWJjc314wbN858\n+OGHxhhjYmNjzb59+875OhQWeqxeqvMyapQx0v/9M2pUiWwWuOKEh4eb8PBwX48B4ATLPQySdPPN\nN0uSQkJCVLNmTUlSaGioCgoK1K5dOy1cuFA7d+5Up06dtHPnTtWpU0flypWTJEVGRmrFihWKiIg4\n43bKly+vo0ePavv27dq/f7+eeuopGWOUl5ennTt3KisrS506dZIkNWzYUA0bNlR6enrRbFu3blVk\nZKQkqV69etqzZ48k6ZprrlGVKlUkSYGBgUWPFxISoqNHj6pDhw5KSEhQZGSkwsLCVLFixXO+Bjk5\neefzUlmKiHDK5QqUx+OQy2UUEZGn7Gz2MNhBWFiIsrNzfT0GTsjI+A9rYkOsib2U9HqEhYWc9brz\nCgaHw3HWyzt06KD+/fsrPz9fAwYM0MGDB7V582bl5+erbNmySk9PL/qrUWfbjiRVr15dVatW1bRp\n0+Tn56fU1FTVr19f27Ztk9vtVt26dZWRkaHly5erefPmOnbsmCSpVq1aysjIUIsWLfTzzz+rUqVK\n53wsY4wk6frrr1dISIgmT56sjh07ns/LUCIiI71KS8uT2x2kiAgORwAALg/nFQznUrlyZQUFBen2\n22+X0+lUhQoV9NxzzykuLk5+fn664YYb1L9/f3366afn3E6FChX0+OOPq3v37vJ6vapWrZqio6PV\nq1cvDRkyRGlpaXI6nRoxYoT8/f21adMmvf/++xo0aJCGDh2q9957Tx6PRyNHjjzn45waEl26dNGI\nESM0evToi30Z/pLISK+iosSeBQDAZcNhTv7IfRF69+6tl19+WdWrVy+JmUrN4sWLtWnTJvXt29fy\ntiW9C47devbDmtgPa2I/rIm92O6QxNkcPXpUXbt2VdOmTS+7WBgzZoxWrVql5ORkX48CAIDtXVQw\nBAQEKCUlpaRmKVUvvPCCr0cAAOCywW8OAmBLnEsCsBeCAQAAWCIYAACAJYIBAABYIhgAAIAlggEA\nAFgiGADY0urVa5WVleXrMQCcQDAAAABLBAMAALBEMAAAAEsEAwAAsEQwAAAASwQDAFviXBKAvRAM\nAADAEsEAAAAsEQwAAMASwQAAACwRDAAAwBLBAMCWOJcEYC8EAwAAsEQwAAAASwQDAACwRDAAAABL\nBAMAALBEMACwJc4lAdgLwQAAACwRDAAAwBLBAAAALBEMAADAEsEAAAAsEQwAbIlzSQD2QjAAAABL\nBAMAALBEMAAAAEsEAwAAsEQwAAAASwQDAFviXBKAvRAMAADAEsEAAAAsuXw9wNUoM9Mpt1uKiHAq\nMtLr63EAALB0RQVDamqqtm7dqhdffFEFBQV64IEHlJSUpPHjx8sYo7y8PL355psKDw/XBx98oEWL\nFsnhcOjBBx/Uo48+WiozZmY61a5doDweyeUKVFpaHtEAALC9KyoYJMnhcBT78+bNmzV69GiFhYUp\nOTlZixcv1v3336/PPvtMH330kYwx6tGjh/72t7+d8wNWFSoEyuXyu+j53G7J4zn+tcfjkNsdpKio\ni94sSkhYWIivR8AJTufx9zJrYj+sib2U1npcccFwkjFGklS5cmUNHz5cQUFB2rNnjxo1aqSNGzdq\n9+7deuyxx2SMUW5urrZv337OYMjJySuRuSIinHK5AuXxOORyGUVE5Ck7mz0MdhAWFqLs7Fxfj4ET\nMjL+w5rYEGtiLyW9HueKjysqGAICApSdnS1JWrt2rSTp1Vdf1RdffKHAwEANHjxYklSjRg3VqVNH\nU6ZMkSRNnz5ddevWLZUZIyO9SkvLk9sdpIgIDkcAAC4PV1QwNGvWTB999JG6d++uBg0aKCQkRG3a\ntFG3bt0UGBioSpUqae/evapXr57uvPNOde3aVQUFBbrttttUpUqVUpszMtKrqCixZwEAcNlwmJP7\n7nFOJb0Ljt169sOa2A9rYj+sib2U5iEJfg8DAACwRDAAAABLBAMAW+JcEoC9EAwAAMASwQAAACwR\nDAAAwBLBAAAALBEMAADAEsEAwJZWr16rrKwsX48B4ASCAQAAWCIYAACAJYIBAABYIhgAAIAlggEA\nAFgiGADYEueSAOyFYAAAAJYIBgAAYIlgAAAAlggGAABgiWAAAACWCAYAtsS5JAB7IRgAAIAlggEA\nAFgiGAAAgCWCAQAAWCIYAACAJYIBgC1xLgnAXggGAABgiWAAAACWCAYAAGCJYAAAAJYIBgAAYIlg\nAGBLnEsCsBeCAQAAWCIYAACAJYIBAABYIhgAAIAlggEAAFgiGADYEueSAOyFYAAAAJZcl/oB0tPT\nFR8fr7feekvR0dFFl7dr104NGjTQqlWr9P/+3/+Tw+HQsWPHdOTIEQ0fPlwNGjTQSy+9pAcffFB/\n+9vfzrjt1NRUvfPOO6pevbqMMXI4HOrRo4cOHDigrVu36sUXXyy6bUJCgrp27aomTZro3Xff1fff\nfy+PxyOn06mBAweqQYMGl/qlAADgsnXJg0GSatasqc8++6woGDZu3Kj8/HxJksPh0HvvvSd/f39J\n0nfffadx48Zp8uTJ57Xttm3bKiEhodhlqampcjgcZ7z9li1btGzZMs2ePVuStH79eg0ePFgff/zx\nBT03AACuBqUSDPXq1VNWVpYOHTqk4OBgpaWlqW3bttq9e7ckyRhTdNvdu3erfPnyp23j9ddf108/\n/aRrr71Wv/zyi5KTk0+77/kIDg7Wb7/9pvnz56tZs2aqV6+e5s2bdxHP7q/LzHTK7ZYiIpyKjPSW\n6mMDAHAhSiUYJKl169b64osv1L59e7ndbvXq1Uu7d++WMUY9e/ZUfn6+9u7dq3vuuUeDBg0qdt8v\nv/xSBw8e1Ny5c7V//3498MADRdctWrRIP/74o4wxuvbaa/X222+fdQaHw6EqVapo0qRJmjlzpiZM\nmKBy5cqpX79+at269SV77qfKzHSqXbtAeTySyxWotLQ8ogEAYHulEgwOh0MxMTF67bXXVK1aNTVp\n0qTYdScPSYwZM0a//PKLKlasWOz+W7ZsUcOGDSVJFStWVI0aNYquO9MhiYCAABUUFBS7LC8vTwEB\nAdqxY4eCgoI0cuRISdJPP/2kJ598UnfeeadCQ0PP+hwqVAiUy+V3YS/AKdxuyeM5/rXH45DbHaSo\nqIveLEpIWFiIr0fACTt2bPf1CDgL3if2UlrrUWp7GKpVq6YjR45o5syZevHFF7Vjx46i604eVujX\nr5/i4+M1a9Ysde/evej6m266SWlpaYqPj9fBgwctT0hTv359TZo0SXl5eQoMDNSBAwe0adMm1a5d\nW999953mzJmjSZMmyd/fX+Hh4QoNDZXTee6/MJKTk3fhT/4UERFOuVyB8ngccrmMIiLylJ3NHgY7\nCAsLUXZ2rq/HwClYE/thTeylpNfjXPFRasEgSdHR0UpLS1N4eHixYDjJ4XAoMTFRjz76aLFDBM2b\nN9c333yjrl27qlKlSipXrpxcrrOPXqNGDXXv3l3dunVTcHCwPB6PXnnlFZUrV06tWrXS1q1b1alT\nJwUFBcnr9WrQoEEKDg6+JM/5zyIjvUpLy5PbHaSICA5HAAAuDw7zVz816ANbt27V+vXrFR0drQMH\nDigmJkZfffVV0d+sKA0lXdRUuv2wJvbDmtgPa2IvV+wehgtVtWpVjR49WjNmzJDX69WAAQNKNRYA\nALjaXRbBUK5cOU2cONHXYwAAcNXiV0MDsCXOJQHYC8EAAAAsEQwAAMASwQAAACwRDAAAwBLBAAAA\nLBEMAGxp9eq1lr8GHkDpIRgAAIAlggEAAFgiGAAAgCWCAQAAWCIYAACAJYIBgC1xLgnAXggGAABg\niWAAAACWCAYAAGCJYAAAAJYIBgAAYIlgAGBLnEsCsBeCAQAAWCIYAACAJYIBAABYIhgAAIAlggEA\nAFgiGADYEueSAOyFYAAAAJYIBgAAYIlgAAAAlggGAABgiWAAAACWCAYAtsS5JAB7IRgAAIAlggEA\nAFgiGAAAgCWCAQAAWCIYAACAJYIBgC1xLgnAXggGAABgyZbBEBcXp23bthW7LD09XQkJCWe9T0FB\ngebNmydJSk1N1VdffXVJZwQA4Gpiy2A4G4fDcdbr9u7dq/nz50uS2rdvr/vuu6+0xgIA4Irn8vUA\nhw4d0tChQ5Wbm6u9e/eqW7dukqSxY8cqJydHAQEBSkpKKnafWbNmacmSJcrPz1eFChU0btw4JScn\na8uWLZo4caK8Xq/CwsL0yCOPKCkpSatXr5bD4VBMTIzi4uL00ksvyd/fX7t27dLvv/+uUaNGqX79\n+qX2nDMznXK7pYgIpyIjvaX2uAAAXCif72HYsWOHYmJiNHXqVE2dOlXTp0+Xw+FQ69atNWPGDDVv\n3lzJycnF7pOTk6MZM2Zozpw5Kiws1Nq1a9W7d2/Vrl1bzzzzTNHtvv76a+3atUtz587VrFmztGjR\nIm3cuFGSVK1aNU2dOlWPPvqo5syZU2rPNzPTqXbtAjV4sNSuXaAyM32+BAAAWPL5HoZrr71WM2bM\n0JIlSxQUFKTCwkJJUpMmTSRJjRo10jfffFPsPmXKlFFCQoLKlSunvXv3yuPxnHHbW7ZsUePGjSVJ\nLpdLERER2rx5syQV7VG47rrr9MMPP1jOWaFCoFwuvwt7kqdwu6WT43o8DrndQYqKuujNooSEhYX4\negScsGPHdl+PgLPgfWIvpbUePg+GadOm6fbbb1dsbKxWrVql5cuXS5LcbrdatmypzMxM1alTp+j2\nGzZs0NKlSzV37lzl5+erQ4cOMsbI6XTK6y2+e7927dpasGCBHnvsMRUWFmrNmjXq0KGDvv3223N+\nHuJMcnLyLv7J6vhhCJcrUB6PQy6XUUREnrKzOSxhB2FhIcrOzvX1GDgFa2I/rIm9lPR6nCs+fB4M\n9913nxLSVwxTAAANYklEQVQTE/Xpp58qNDRULpdLBQUFWrp0qaZPn66QkBAlJSXp559/liTdeOON\nCgwMVLdu3WSMUeXKlbV37141bNhQhYWFevPNNxUQECBJuvfee7Vy5UrFxsaqsLBQ0dHRpfpZhTOJ\njPQqLS1PbneQIiLy+AwDAOCy4DDGGF8PcTko6aKm0u2HNbEf1sR+WBN7Kc09DHziDgAAWCIYAACA\nJYIBgC1xLgnAXggGAABgiWAAAACWCAYAAGCJYAAAAJYIBgAAYIlgAGBLq1evVVZWlq/HAHACwQAA\nACwRDAAAwBLBAAAALBEMAADAEsEAAAAsEQwAbIlzSQD2QjAAAABLBAMAALBEMAAAAEsEAwAAsEQw\nAAAASwQDAFviXBKAvRAMAADAEsEAAAAsEQwAAMASwQAAACwRDAAAwBLBAMCWOJcEYC8EAwAAsEQw\nAAAASwQDAACwRDAAAABLBAMAALBEMACwJc4lAdgLwQAAACwRDAAAwBLBAAAALBEMAADAEsEAAAAs\nEQwAbIlzSQD2QjAAAABLV2wwfPvtt5o3b57l7bZu3aq4uLhSmAgAgMuXy9cDXCrNmjU779s6HI5L\nOAkAAJe/KzYYUlNT9e2332rXrl2aM2eOJOmRRx7RmDFj5O/vr/79+0uSKlWqVOqzZWY65XZLERFO\nRUZ6S/3xAQD4q67YQxInnbr34OTXkydPVkxMjGbMmKGWLVuW6jyZmU61axeowYOldu0ClZl5xS8B\nAOAKcMXuYTgTr/f4T/NZWVnq0qWLJKlx48aaPXu25X0rVAiUy+V30TO43ZLHc/xrj8chtztIUVEX\nvVmUkLCwEF+PgBN27Nju6xFwFrxP7KW01uOKDoaQkBDt27dPxhjl5ubql19+kSTVrl1ba9asUd26\ndeV2u89rWzk5eSUyU0SEUy5XoDweh1wuo4iIPGVnc1jCDsLCQpSdnevrMXAK1sR+WBN7Ken1OFd8\nXNHBUL58eTVt2lQdO3ZU9erVFR4eLknq3bu3+vfvr88++0zVqlUr1ZkiI71KS8uT2x2kiIg8PsMA\nALgsOIwxxtdDXA5KuqipdPthTeyHNbEf1sReSnMPA5+4AwAAlggGAABgiWAAYEucSwKwF4IBAABY\nIhgAAIAlggEAAFgiGAAAgCWCAQAAWCIYANjS6tVrlZWV5esxAJxAMAAAAEsEAwAAsEQwAAAASwQD\nAACwRDAAAABLBAMAW+JcEoC9EAwAAMASwQAAACwRDAAAwBLBAAAALBEMAADAEsEAwJY4lwRgLwQD\nAACwRDAAAABLBAMAALBEMAAAAEsEAwAAsEQwALAlziUB2AvBAAAALBEMAADAEsEAAAAsEQwAAMAS\nwQAAACwRDABsiXNJAPZCMAAAAEsEAwAAsEQwAAAASwQDAACwRDAAAABLBAMAW+JcEoC9EAwAAMDS\nJQ+GgwcPatGiRRe9nblz5+rYsWMlMFFxGzduVGZmZolvFwCAK8klD4b169dr2bJlF72dyZMnX5Jg\nWLJkiTZv3lzi2wUA4ErisrpBamqqli9frvz8fO3cuVNPPvmk6tWrp+HDh8vPz08BAQFKTEzUdddd\nd8b7Jycna8OGDZo3b55++OEH5eTk6ODBg3r33Xc1ZcoUrV69WseOHVOPHj3Upk0bZWRkaPz48TLG\nKC8vT2+++aYyMjL0+++/KyEhQfHx8UpOTlaZMmW0Z88ePfLII1q5cqU2bNig+Ph4xcbGKj09XW+/\n/bb8/Px0ww036PXXX9fChQuLPY+nnnpKd911l1JSUlSmTBk1aNBAt956a4m/wAAAXAqZmU653VJE\nhFORkd5L/niWwSBJhw4d0r/+9S9t375dvXv3VlBQkEaMGKG6devqyy+/1MiRI/XOO++c8b69e/fW\nnDlz1LlzZ/3www+666679Nhjj+mbb77Rrl27NGvWLBUUFKhLly66++67tWnTJo0ePVphYWFKTk7W\n4sWL9fTTT2vSpEkaM2aM1qxZo7179+qTTz7Rf/7zH/Xr109Lly7Vr7/+qr59+yo2NlavvPKKPvro\nI1WsWFFjx45VamqqXC7Xac/j4YcfVocOHRQWFkYsAAAuG5mZTrVrFyiPR3K5ApWWlnfJo+G8gqF+\n/fqSpKpVq+ro0aM6fPiw6tatK0lq0qSJ3nrrrfN+wBo1akg6/tmBtWvXKj4+XsYYHTt2TL/88ouq\nVKmi4cOHKygoSHv27FGjRo0kScYYGWMkSXXq1JHT6VRISIiqV68uPz8/lS9fXkePHtX+/fuVnZ2t\nfv36yRijgoICNW3aVDfccEOx51FQUHDeM0tShQqBcrn8/tJ9rISFhZTo9nDxWBP72LFju69HwFnw\nPvE9t1vyeI5/7fE45HYHKSrq0j7meQWDw+Eo9ufKlStrw4YNqlu3rtLT08/5V5+cTqe8Xm+xP0tS\nzZo1dccdd2jYsGEyxmjixImqXr26nnjiCS1dulSBgYEaPHhw0f38/PyKtnPqPCcj4qSKFSuqatWq\nmjhxooKDg7Vs2TIFBQVp9+7dZ7yfw+E4r89G5OTkWd7mrwgLC1F2dm6JbhMXhzWxH9bEflgTe4iI\ncMrlCpTH45DLZRQRkafs7Ivfw3CuGDyvYDiVw+FQYmKihg8fLun4N/IRI0ac9fbVq1fXxo0b9f77\n7xe7vEWLFkpPT1f37t115MgR3X///QoKCtJDDz2kbt26KTAwUJUqVdLevXslSY0bN1avXr307LPP\nnjbPnw0ZMkS9evWS1+tVSEiIkpKStHv37jPe75ZbbtEbb7yh2rVr67/+67/+6ssBAECpi4z0Ki0t\nT253kCIiLv3hCElymD//iI4zKumiptLthzWxH9bEflgTeynp9SjRPQxn07dvXx08eLDoz8YYhYaG\nasKECSX1EAAAwEdKLBjGjRtXUpsCAAA2w6+GBmBLnEsCsBeCAQAAWCIYAACAJYIBAABYIhgAAIAl\nggEAAFgiGADY0urVa5WVleXrMQCcQDAAAABLBAMAALBEMAAAAEsEAwAAsEQwAAAASwQDAFviXBKA\nvRAMAADAEsEAAAAsEQwAAMASwQAAACwRDAAAwJLDGGN8PQQAALA39jAAAABLBAMAALBEMAAAAEsE\nAwAAsEQwAAAASwQDAACwRDAAAABLBEMp++KLL/Tiiy+e8bq5c+eqY8eOio2N1ddff126g12Fjh49\nqueee07du3fX008/rZycnNNuM2LECHXs2FHx8fGKj4/XoUOHfDDplc0Yo9dee02xsbGKj4/Xzp07\ni12/bNkyderUSbGxsZo3b56Ppry6WK3J9OnTFRMTU/S+yMrK8s2gV5kff/xRcXFxp11eau8Rg1KT\nmJhooqKiTEJCwmnXZWdnm5iYGFNYWGhyc3NNTEyMKSgo8MGUV49p06aZcePGGWOM+fTTT01iYuJp\nt+natavJyckp7dGuKkuWLDGDBw82xhjz73//2/z9738vuq6wsNC0atXK5ObmmoKCAtOxY0ezb98+\nX4161TjXmhhjTP/+/c1PP/3ki9GuWlOmTDExMTHmkUceKXZ5ab5H2MNQiho1aqT//u//PuN1brdb\njRs3lsvlUnBwsG688UZt2LChdAe8yqxevVr33HOPJOmee+7R999/X+x6Y4y2b9+uV199VV27dtWC\nBQt8MeYVb/Xq1WrWrJkk6bbbbtPatWuLrtuyZYvCw8MVHBwsf39/NW7cWBkZGb4a9apxrjWRpJ9+\n+knJycnq1q2b3n33XV+MeNUJDw/XhAkTTru8NN8jrkuy1avc/PnzNWPGjGKX/eMf/1BUVJTS09PP\neJ9Dhw4pJCSk6M+BgYHKzc29pHNeTc60JpUqVVJwcLAkKSgo6LTDDXl5eYqLi1OPHj3k8XgUHx+v\nW2+9VTfddFOpzX01+PN/+y6XS16vV06n87TrgoKCeF+UgnOtiSQ9+OCD6t69u4KDg9WnTx8tX75c\n9957r6/GvSq0atVKu3btOu3y0nyPEAyXQKdOndSpU6e/dJ/g4OBi37AOHz6s0NDQkh7tqnWmNenb\nt68OHz4s6fjrfeqbTpLKlSunuLg4BQQEKCAgQHfeeafWr19PMJSw4ODgonWQVOwbE+8L3zjXmkjS\nY489VhTb9957r9atW0cw+Ehpvkc4JGETERERWr16tQoKCpSbm6utW7eqTp06vh7ritaoUSMtX75c\nkrR8+XJFRkYWu37btm3q2rWrjDEqLCzU6tWr1aBBA1+MekU7dR3+/e9/FwuyWrVqafv27frjjz9U\nUFCgjIwMNWzY0FejXjXOtSaHDh1STEyMjhw5ImOMVq5cyfuiFJk/nS+yNN8j7GHwsenTpys8PFz3\n3Xef4uLi1K1bNxljlJCQoDJlyvh6vCta165dNWjQIHXr1k1lypTRm2++Kan4mjz88MPq3Lmz/P39\n1b59e9WqVcvHU195WrVqpRUrVig2NlbS8cN3ixYt0pEjR9S5c2e99NJLeuKJJ2SMUefOnVW5cmUf\nT3zls1qThISEor1vd911V9FngXDpORwOSfLJe4TTWwMAAEsckgAAAJYIBgAAYIlgAAAAlggGAABg\niWAAAACWCAYAAGCJYAAAAJb+P5uinxGo3QJ7AAAAAElFTkSuQmCC\n",
"text/plain": [
"<matplotlib.figure.Figure at 0x116f34208>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pm.forestplot(trace_med_manage, vars=['μ'], ylabels=plot_labels)"
]
},
{
"cell_type": "code",
"execution_count": 80,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": [
"def get_data(intervention):\n",
" \n",
" intervention_data = dataset[(dataset.intervention_cat==intervention)\n",
" & ~dataset[outcome_cats].isnull().sum(axis=1).astype(bool)]\n",
" \n",
" followup_masked = masked_values(intervention_data.followup_interval.values, np.nan)\n",
" followup_min, followup_max = intervention_data[['fup_min', 'fup_max']].values.T\n",
"\n",
" outcomes = intervention_data[[ 'hysterectomy', 'myomectomy', 'uae',\n",
" 'MRIgFUS', 'ablation', 'iud', 'no_treatment']].values\n",
" \n",
" if np.isnan(outcomes).any():\n",
" print('Missing values in outcomes for', intervention)\n",
"\n",
" followup_n = intervention_data.followup_n.values\n",
"\n",
" # Center age at 40\n",
" age_masked = masked_values(intervention_data['BL Mean'].values - 40, np.nan)\n",
" \n",
" studies = intervention_data.study_id.unique()\n",
" study_index = np.array([np.argwhere(studies==i).squeeze() for i in intervention_data.study_id])\n",
"\n",
" study_id = intervention_data.study_id.values\n",
" \n",
" n_studies = len(set(study_id))\n",
"\n",
" n_outcomes = 7\n",
" arms = len(outcomes) \n",
" intervention_data = dataset[(dataset.intervention_cat==intervention)\n",
" & ~dataset[outcome_cats].isnull().sum(axis=1).astype(bool)]\n",
" \n",
" followup_masked = masked_values(intervention_data.followup_interval.values, 17.33)\n",
" followup_min, followup_max = intervention_data[['fup_min', 'fup_max']].values.T\n",
"\n",
" outcomes = intervention_data[[ 'hysterectomy', 'myomectomy', 'uae',\n",
" 'MRIgFUS', 'ablation', 'iud', 'no_treatment']].values\n",
"\n",
" followup_n = intervention_data.followup_n.values\n",
"\n",
" age_masked = masked_values(intervention_data['BL Mean'].values, 41.33)\n",
" \n",
" studies = intervention_data.study_id.unique()\n",
" study_index = np.array([np.argwhere(studies==i).squeeze() for i in intervention_data.study_id])\n",
"\n",
" study_id = intervention_data.study_id.values\n",
" \n",
" n_studies = len(set(study_id))\n",
"\n",
" n_outcomes = 7\n",
" arms = len(outcomes)\n",
" \n",
" return locals()"
]
},
{
"cell_type": "code",
"execution_count": 81,
"metadata": {
"collapsed": false
},
"outputs": [],
"source": [
"foo = get_data('med_manage')"
]
},
{
"cell_type": "code",
"execution_count": 84,
"metadata": {
"collapsed": false
},
"outputs": [
{
"data": {
"text/plain": [
"masked_array(data = [2.5 4.5 3.6000000000000014 -- 1.6000000000000014 1.6000000000000014\n",
" 1.1000000000000014 1.1000000000000014 -1 -1 -1 -0.3999999999999986\n",
" -0.3999999999999986 -0.3999999999999986 1.6000000000000014\n",
" 1.6000000000000014 1.6000000000000014 1.6000000000000014\n",
" 1.6000000000000014 1.6000000000000014 1.6000000000000014\n",
" 1.6000000000000014 1.2999999999999972 1.2999999999999972\n",
" 1.2999999999999972 1.2999999999999972 1.1000000000000014\n",
" 1.1000000000000014 1.1000000000000014 1.1000000000000014\n",
" -0.3299999999999983 -3.1300000000000026 -9.059999999999999 -9\n",
" 16.200000000000003 10.200000000000003 10.600000000000001\n",
" -5.600000000000001 1 4 2.200000000000003 2.200000000000003 2 2 --\n",
" 9.200000000000003 -7.399999999999999 -- 9.200000000000003\n",
" -7.399999999999999 -0.5 -1.8999999999999986 -3 -3 -- -- -- -- -- --],\n",
" mask = [False False False True False False False False False False False False\n",
" False False False False False False False False False False False False\n",
" False False False False False False False False False False False False\n",
" False False False False False False False False True False False True\n",
" False False False False False False True True True True True True],\n",
" fill_value = ?)"
]
},
"execution_count": 84,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"foo['age_masked'] - 40"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {
"collapsed": true
},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.5.1"
}
},
"nbformat": 4,
"nbformat_minor": 0
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment