Skip to content

Instantly share code, notes, and snippets.

@borgeslt

borgeslt/IGTI_Trabalho_prático_1.ipynb

Created July 20, 2020 11:25
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save borgeslt/0ef5e7a9f30674a7f796ef74e3e4e0e0 to your computer and use it in GitHub Desktop.
Save borgeslt/0ef5e7a9f30674a7f796ef74e3e4e0e0 to your computer and use it in GitHub Desktop.
Download ZIP
Display the source blob
Display the rendered blob
Raw
IGTI_Trabalho_prático_1.ipynb
{
"nbformat": 4,
"nbformat_minor": 0,
"metadata": {
"colab": {
"name": "IGTI_Trabalho_prático_1.ipynb",
"provenance": [],
"authorship_tag": "ABX9TyMSLLSxKQmpT4qAmTSgXvNz"
},
"kernelspec": {
"name": "python3",
"display_name": "Python 3"
}
},
"cells": [
{
"cell_type": "markdown",
"metadata": {
"id": "I9W-ehIbbUX7",
"colab_type": "text"
},
"source": [
"# *Bootcamp* IGTI - Analista de *Machine Learning*: Projeto Prático 1 -- Fundamentos\n",
"\n",
"Aplicação dos conceitos de análise e modelamento de *Machine Learning* aprendidos no Módulo 1 do Bootcamp.\n",
"\n",
"\n",
"**Objetivos:**\n",
"* Conhecimento do dataset\n",
"* Limpeza dos dados\n",
"* Identificação de *Outliers*\n",
"* Análise de regressão linear\n",
"\n",
"\n",
"Para qualquer aplicação que utilize algoritmos de *Machine Learning*, precisamos realizar 7 etapas básicas:\n",
"\n",
"* Coleta de dados\n",
"* Preparação dos dados\n",
"* Treinamento do modelo\n",
"* Avaliação do modelo\n",
"* Sintonia dos parâmetros\n",
"* Previsão\n"
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "ZoxNm0NxyeXT",
"colab_type": "text"
},
"source": [
"Mas antes de começar, vamos entender...\n",
"\n",
"> ### **O que é Regressão Linear?**\n",
"\n",
"É um dos algoritmos de *Machine Learning* mais conhecidos, ele é utilizado para estimar valores reais baseado na relação entre variáveis dependentes e independentes contínuas. \n",
"\n",
"Essa relação pode ser traduzida para uma equação matemática que tem como saída uma linha na qual vamos ajustando nosso parâmetros, para chegar o mais próximo possível dessa linha com os resultados do modelo de ML criado. \n",
"\n",
"Essa linha é conhecida como **Linha de Regressão** e é representado por uma equação linear:\n",
"\n",
"<p align=\"center\"> Y = a * x + b </p>\n",
"\n",
"Onde:\n",
"* Y - Variável Dependente\n",
"* a - Coeficiente Angular\n",
"* x - Variável Independente\n",
"* b - Intercepção\n",
"\n",
"Os coeficientes a e b são derivados baseados na minimização da soma dos quadrados da diferença da distância entre os pontos da regressão linear.\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "DvtuDkI7c6z_",
"colab_type": "code",
"colab": {}
},
"source": [
"# importar bibliotecas\n",
"import pandas as pd\n",
"import numpy as np\n",
"import matplotlib.pyplot as plt\n"
],
"execution_count": 13,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "5UBSMbtPdPoc",
"colab_type": "code",
"colab": {
"resources": {
"http://localhost:8080/nbextensions/google.colab/files.js": {
"data": "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",
"ok": true,
"headers": [
[
"content-type",
"application/javascript"
]
],
"status": 200,
"status_text": ""
}
},
"base_uri": "https://localhost:8080/",
"height": 76
},
"outputId": "46453f6c-ed62-45c6-f182-21bce1e630fa"
},
"source": [
"# upload do aquivo data.csv\n",
"from google.colab import files\n",
"uploaded = files.upload()"
],
"execution_count": 14,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/html": [
"\n",
" <input type=\"file\" id=\"files-b0eb8692-87dc-4169-a94f-6afb07009eed\" name=\"files[]\" multiple disabled\n",
" style=\"border:none\" />\n",
" <output id=\"result-b0eb8692-87dc-4169-a94f-6afb07009eed\">\n",
" Upload widget is only available when the cell has been executed in the\n",
" current browser session. Please rerun this cell to enable.\n",
" </output>\n",
" <script src=\"/nbextensions/google.colab/files.js\"></script> "
],
"text/plain": [
"<IPython.core.display.HTML object>"
]
},
"metadata": {
"tags": []
}
},
{
"output_type": "stream",
"text": [
"Saving data.csv to data (1).csv\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "WmGYOOVKddCH",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 230
},
"outputId": "c0113a76-0027-48a8-bf47-c959b5e935b1"
},
"source": [
"# importar o dataframe\n",
"df = pd.read_csv(\"data.csv\")\n",
"\n",
"# visualizar as primeiras entradas do dataset\n",
"df.head()"
],
"execution_count": 15,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>valid_import</th>\n",
" <th>item</th>\n",
" <th>importer_id</th>\n",
" <th>exporter_id</th>\n",
" <th>country_of_origin</th>\n",
" <th>declared_quantity</th>\n",
" <th>declared_cost</th>\n",
" <th>mode_of_transport</th>\n",
" <th>route</th>\n",
" <th>date_of_departure</th>\n",
" <th>date_of_arrival</th>\n",
" <th>declared_weight</th>\n",
" <th>actual_weight</th>\n",
" <th>days_in_transit</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>True</td>\n",
" <td>cigar</td>\n",
" <td>111</td>\n",
" <td>222</td>\n",
" <td>India</td>\n",
" <td>129</td>\n",
" <td>3784.402551</td>\n",
" <td>sea</td>\n",
" <td>asia</td>\n",
" <td>04/25/2019</td>\n",
" <td>05/13/2019</td>\n",
" <td>1608.605135</td>\n",
" <td>1637.661221</td>\n",
" <td>18.232857</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>True</td>\n",
" <td>cigar</td>\n",
" <td>111</td>\n",
" <td>222</td>\n",
" <td>India</td>\n",
" <td>104</td>\n",
" <td>3081.350806</td>\n",
" <td>sea</td>\n",
" <td>america</td>\n",
" <td>04/22/2019</td>\n",
" <td>05/24/2019</td>\n",
" <td>831.719301</td>\n",
" <td>848.273419</td>\n",
" <td>32.436029</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>True</td>\n",
" <td>cigar</td>\n",
" <td>111</td>\n",
" <td>222</td>\n",
" <td>India</td>\n",
" <td>130</td>\n",
" <td>4414.125741</td>\n",
" <td>sea</td>\n",
" <td>europe</td>\n",
" <td>04/29/2019</td>\n",
" <td>05/16/2019</td>\n",
" <td>1527.704165</td>\n",
" <td>1582.063911</td>\n",
" <td>16.996206</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>True</td>\n",
" <td>cigar</td>\n",
" <td>111</td>\n",
" <td>222</td>\n",
" <td>India</td>\n",
" <td>143</td>\n",
" <td>2533.535991</td>\n",
" <td>sea</td>\n",
" <td>panama</td>\n",
" <td>05/05/2019</td>\n",
" <td>05/25/2019</td>\n",
" <td>1138.680563</td>\n",
" <td>1179.993817</td>\n",
" <td>19.965886</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>True</td>\n",
" <td>cigar</td>\n",
" <td>111</td>\n",
" <td>222</td>\n",
" <td>China</td>\n",
" <td>141</td>\n",
" <td>4396.397887</td>\n",
" <td>sea</td>\n",
" <td>asia</td>\n",
" <td>05/14/2019</td>\n",
" <td>06/05/2019</td>\n",
" <td>761.744581</td>\n",
" <td>781.735080</td>\n",
" <td>22.160034</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" valid_import item ... actual_weight days_in_transit\n",
"0 True cigar ... 1637.661221 18.232857\n",
"1 True cigar ... 848.273419 32.436029\n",
"2 True cigar ... 1582.063911 16.996206\n",
"3 True cigar ... 1179.993817 19.965886\n",
"4 True cigar ... 781.735080 22.160034\n",
"\n",
"[5 rows x 14 columns]"
]
},
"metadata": {
"tags": []
},
"execution_count": 15
}
]
},
{
"cell_type": "code",
"metadata": {
"id": "yVUhl6JFdtWg",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 395
},
"outputId": "f98d0983-4d23-4ade-e40c-e71e9ff3f973"
},
"source": [
"# ver as características do dataset\n",
"df.info()"
],
"execution_count": 16,
"outputs": [
{
"output_type": "stream",
"text": [
"<class 'pandas.core.frame.DataFrame'>\n",
"RangeIndex: 120 entries, 0 to 119\n",
"Data columns (total 14 columns):\n",
" # Column Non-Null Count Dtype \n",
"--- ------ -------------- ----- \n",
" 0 valid_import 120 non-null bool \n",
" 1 item 120 non-null object \n",
" 2 importer_id 120 non-null int64 \n",
" 3 exporter_id 120 non-null int64 \n",
" 4 country_of_origin 120 non-null object \n",
" 5 declared_quantity 120 non-null int64 \n",
" 6 declared_cost 120 non-null float64\n",
" 7 mode_of_transport 120 non-null object \n",
" 8 route 120 non-null object \n",
" 9 date_of_departure 120 non-null object \n",
" 10 date_of_arrival 120 non-null object \n",
" 11 declared_weight 120 non-null float64\n",
" 12 actual_weight 120 non-null float64\n",
" 13 days_in_transit 120 non-null float64\n",
"dtypes: bool(1), float64(4), int64(3), object(6)\n",
"memory usage: 12.4+ KB\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "yLf18PbLm0pw",
"colab_type": "text"
},
"source": [
"### **Quantas colunas e linhas existem no *dataset***"
]
},
{
"cell_type": "code",
"metadata": {
"id": "TYAnb3aEnCxP",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 53
},
"outputId": "71028984-dbb9-48e1-ea90-8b8ed58db82f"
},
"source": [
"print(\"Número de Linhas:\", df.shape[0])\n",
"print(\"Número de Colunas:\", df.shape[1])"
],
"execution_count": 17,
"outputs": [
{
"output_type": "stream",
"text": [
"Número de Linhas: 120\n",
"Número de Colunas: 14\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "3Ii8nMvGk2c_",
"colab_type": "text"
},
"source": [
"### **Vamos analisar se existem colunas com valores nulos**"
]
},
{
"cell_type": "code",
"metadata": {
"id": "AaxhVlfYlMOa",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 287
},
"outputId": "647873b4-8bd3-4968-9b68-b522df022cc3"
},
"source": [
"df.isnull().sum()"
],
"execution_count": 18,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"valid_import 0\n",
"item 0\n",
"importer_id 0\n",
"exporter_id 0\n",
"country_of_origin 0\n",
"declared_quantity 0\n",
"declared_cost 0\n",
"mode_of_transport 0\n",
"route 0\n",
"date_of_departure 0\n",
"date_of_arrival 0\n",
"declared_weight 0\n",
"actual_weight 0\n",
"days_in_transit 0\n",
"dtype: int64"
]
},
"metadata": {
"tags": []
},
"execution_count": 18
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "k0LNJy7zlPJ-",
"colab_type": "text"
},
"source": [
"### **Vamos analisar as estatísticas do *Dataset***"
]
},
{
"cell_type": "code",
"metadata": {
"id": "_7wthDB3l-A8",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 326
},
"outputId": "2e30b5ab-d0a7-4135-dd13-6f106d178578"
},
"source": [
"df.describe()"
],
"execution_count": 19,
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>importer_id</th>\n",
" <th>exporter_id</th>\n",
" <th>declared_quantity</th>\n",
" <th>declared_cost</th>\n",
" <th>declared_weight</th>\n",
" <th>actual_weight</th>\n",
" <th>days_in_transit</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>count</th>\n",
" <td>120.0</td>\n",
" <td>120.0</td>\n",
" <td>120.000000</td>\n",
" <td>120.000000</td>\n",
" <td>120.000000</td>\n",
" <td>120.000000</td>\n",
" <td>120.000000</td>\n",
" </tr>\n",
" <tr>\n",
" <th>mean</th>\n",
" <td>111.0</td>\n",
" <td>222.0</td>\n",
" <td>127.458333</td>\n",
" <td>6743.649881</td>\n",
" <td>1264.702934</td>\n",
" <td>1306.429806</td>\n",
" <td>35.424705</td>\n",
" </tr>\n",
" <tr>\n",
" <th>std</th>\n",
" <td>0.0</td>\n",
" <td>0.0</td>\n",
" <td>14.641311</td>\n",
" <td>2991.797050</td>\n",
" <td>633.149971</td>\n",
" <td>656.911704</td>\n",
" <td>26.571591</td>\n",
" </tr>\n",
" <tr>\n",
" <th>min</th>\n",
" <td>111.0</td>\n",
" <td>222.0</td>\n",
" <td>100.000000</td>\n",
" <td>1441.012419</td>\n",
" <td>18.459509</td>\n",
" <td>19.275241</td>\n",
" <td>12.410325</td>\n",
" </tr>\n",
" <tr>\n",
" <th>25%</th>\n",
" <td>111.0</td>\n",
" <td>222.0</td>\n",
" <td>115.750000</td>\n",
" <td>4442.903914</td>\n",
" <td>820.314400</td>\n",
" <td>841.763738</td>\n",
" <td>18.225625</td>\n",
" </tr>\n",
" <tr>\n",
" <th>50%</th>\n",
" <td>111.0</td>\n",
" <td>222.0</td>\n",
" <td>131.500000</td>\n",
" <td>6010.218745</td>\n",
" <td>1255.597743</td>\n",
" <td>1305.716419</td>\n",
" <td>27.044293</td>\n",
" </tr>\n",
" <tr>\n",
" <th>75%</th>\n",
" <td>111.0</td>\n",
" <td>222.0</td>\n",
" <td>139.000000</td>\n",
" <td>8887.095370</td>\n",
" <td>1711.314045</td>\n",
" <td>1763.681083</td>\n",
" <td>44.356374</td>\n",
" </tr>\n",
" <tr>\n",
" <th>max</th>\n",
" <td>111.0</td>\n",
" <td>222.0</td>\n",
" <td>149.000000</td>\n",
" <td>14281.325362</td>\n",
" <td>2806.338955</td>\n",
" <td>2918.681683</td>\n",
" <td>147.787560</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" importer_id exporter_id ... actual_weight days_in_transit\n",
"count 120.0 120.0 ... 120.000000 120.000000\n",
"mean 111.0 222.0 ... 1306.429806 35.424705\n",
"std 0.0 0.0 ... 656.911704 26.571591\n",
"min 111.0 222.0 ... 19.275241 12.410325\n",
"25% 111.0 222.0 ... 841.763738 18.225625\n",
"50% 111.0 222.0 ... 1305.716419 27.044293\n",
"75% 111.0 222.0 ... 1763.681083 44.356374\n",
"max 111.0 222.0 ... 2918.681683 147.787560\n",
"\n",
"[8 rows x 7 columns]"
]
},
"metadata": {
"tags": []
},
"execution_count": 19
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "4-7nemEm4K-j",
"colab_type": "text"
},
"source": [
"Pela tabela podemos encontrar, facilmente, informações importantes como a média e o desvio-padrão das variáveis."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "EObtsGfFnNx9",
"colab_type": "text"
},
"source": [
"### **Existem *outliers* nas variáveis ``declared_quantity`` e ``days_in_transit``?**"
]
},
{
"cell_type": "code",
"metadata": {
"id": "5ukSxC7rnWQ3",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 266
},
"outputId": "62a58d1f-531a-46eb-bbfe-4f2704217649"
},
"source": [
"# identificar possíveis outliers\n",
"df[['declared_quantity', 'days_in_transit']].boxplot();"
],
"execution_count": 20,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 432x288 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "BKEia4Kl34El",
"colab_type": "text"
},
"source": [
"A partir dos gráficos, chegamos a conclusão que na variável ```days_in_transit``` existem possíveis *outliers*."
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "DaPGIVkhpsiv",
"colab_type": "text"
},
"source": [
"## Aplicando o modelo de *Machine Learning*\n",
"\n",
"\n"
]
},
{
"cell_type": "code",
"metadata": {
"id": "qRvMpmNCn8Le",
"colab_type": "code",
"colab": {}
},
"source": [
"# realizando a análise da regressão\n",
"x = df.declared_weight.values # variável independente\n",
"y = df.actual_weight # variável dependente"
],
"execution_count": 21,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "Al_cgSH4o5yj",
"colab_type": "code",
"colab": {}
},
"source": [
"# importar o modelo de regressão linear univariada\n",
"from sklearn.linear_model import LinearRegression"
],
"execution_count": 22,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "6XVaDrJ6pERA",
"colab_type": "code",
"colab": {}
},
"source": [
"# realiza a contrução do modelo de regressão\n",
"reg = LinearRegression()\n",
"x_reshapeed = x.reshape((-1, 1)) # transforma os dados para o plano 2D\n",
"regressao = reg.fit (x_reshapeed, y) #encontra os coeficientes (realiza a regressão)"
],
"execution_count": 23,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "a1G35DZQpgWj",
"colab_type": "code",
"colab": {}
},
"source": [
"# realiza a previsão\n",
"previsao = reg.predict(x_reshapeed)"
],
"execution_count": 24,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "8dBM9WUWpndu",
"colab_type": "code",
"colab": {}
},
"source": [
"# análise do modelo\n",
"from sklearn.metrics import r2_score"
],
"execution_count": 25,
"outputs": []
},
{
"cell_type": "code",
"metadata": {
"id": "LYv3thynp3ON",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 53
},
"outputId": "98a8cfcb-de4f-40e6-c74a-4dc9aa25fb4c"
},
"source": [
"# parâmetros encontrados\n",
"print('Y = {}X {}'.format(reg.coef_, reg.intercept_))\n",
"\n",
"R_2 = r2_score(y, previsao) # calcula o R2\n",
"\n",
"print(\"Coeficiente de Correlação (R2):\", R_2)"
],
"execution_count": 26,
"outputs": [
{
"output_type": "stream",
"text": [
"Y = [1.03718115]X -5.296233030439225\n",
"Coeficiente de Correlação (R2): 0.9993288165644932\n"
],
"name": "stdout"
}
]
},
{
"cell_type": "markdown",
"metadata": {
"id": "5qWPvtSfqOrm",
"colab_type": "text"
},
"source": [
"### **Pelo Coeficiente de Correlação (R2), o que é possível afirmar sobre a relação entre as variáveis?**\n",
"\n",
"Podemos afirmar que a análise possui um bom 'fit' dos dados, ou seja, é possível prever o peso real de um indivíduo a partir do seu peso declarado."
]
},
{
"cell_type": "code",
"metadata": {
"id": "JEyMOqCorLlP",
"colab_type": "code",
"colab": {
"base_uri": "https://localhost:8080/",
"height": 382
},
"outputId": "def1b679-0d30-41c7-b008-ea6611fc399f"
},
"source": [
"# plotar o gráfico dos dados\n",
"plt.figure(figsize=(4, 4), dpi=100)\n",
"plt.scatter(x, y, color='red') # plota o gráfio de dispersão\n",
"plt.plot(x, previsao, color='black', linewidth = 2) # plota a linha do gráfico\n",
"plt.xlabel(\"Peso Declarado\")\n",
"plt.ylabel(\"Peso Real\")\n",
"plt.show();"
],
"execution_count": 27,
"outputs": [
{
"output_type": "display_data",
"data": {
"image/png": "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\n",
"text/plain": [
"<Figure size 400x400 with 1 Axes>"
]
},
"metadata": {
"tags": [],
"needs_background": "light"
}
}
]
}
]
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment