{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np \n",
"import matplotlib.pyplot as plt \n",
"import seaborn as sns\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.preprocessing import OneHotEncoder\n",
"from sklearn.preprocessing import LabelEncoder\n",
"from sklearn.metrics import accuracy_score, f1_score, precision_score, recall_score, roc_auc_score\n",
"\n",
"from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier, AdaBoostClassifier\n",
"from sklearn.tree import DecisionTreeClassifier\n",
"from sklearn.neighbors import KNeighborsClassifier\n",
"from xgboost import XGBClassifier\n",
"from catboost import CatBoostClassifier\n",
"\n",
"from sklearn.compose import ColumnTransformer\n",
"from sklearn.pipeline import Pipeline\n",
"from sklearn.preprocessing import StandardScaler\n",
"\n",
"import warnings\n",
"\n",
"# Ignore warnings\n",
"warnings.filterwarnings(\"ignore\")"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"os.chdir(\"/config/workspace\")"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" Temparature | \n",
" Humidity | \n",
" Moisture | \n",
" Soil Type | \n",
" Crop Type | \n",
" Nitrogen | \n",
" Potassium | \n",
" Phosphorous | \n",
" Fertilizer Name | \n",
"
\n",
" \n",
" \n",
" \n",
" | 0 | \n",
" 26 | \n",
" 52 | \n",
" 38 | \n",
" Sandy | \n",
" Maize | \n",
" 37 | \n",
" 0 | \n",
" 0 | \n",
" Urea | \n",
"
\n",
" \n",
" | 1 | \n",
" 29 | \n",
" 52 | \n",
" 45 | \n",
" Loamy | \n",
" Sugarcane | \n",
" 12 | \n",
" 0 | \n",
" 36 | \n",
" DAP | \n",
"
\n",
" \n",
" | 2 | \n",
" 34 | \n",
" 65 | \n",
" 62 | \n",
" Black | \n",
" Cotton | \n",
" 7 | \n",
" 9 | \n",
" 30 | \n",
" 14-35-14 | \n",
"
\n",
" \n",
" | 3 | \n",
" 32 | \n",
" 62 | \n",
" 34 | \n",
" Red | \n",
" Tobacco | \n",
" 22 | \n",
" 0 | \n",
" 20 | \n",
" 28-28 | \n",
"
\n",
" \n",
" | 4 | \n",
" 28 | \n",
" 54 | \n",
" 46 | \n",
" Clayey | \n",
" Paddy | \n",
" 35 | \n",
" 0 | \n",
" 0 | \n",
" Urea | \n",
"
\n",
" \n",
"
\n",
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# checking the data distribution \n",
"plt.figure(figsize=(10, 7))\n",
"\n",
"sns.countplot(data=df, x='Fertilizer Name')\n",
"plt.title(\"Dataset Distribution\")\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array(['Sandy', 'Loamy', 'Black', 'Red', 'Clayey'], dtype=object)"
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# soil type\n",
"df['Soil Type'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"array(['Maize', 'Sugarcane', 'Cotton', 'Tobacco', 'Paddy', 'Barley',\n",
" 'Wheat', 'Millets', 'Oil seeds', 'Pulses', 'Ground Nuts'],\n",
" dtype=object)"
]
},
"execution_count": 9,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"df['Crop Type'].unique()"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Shape of the training dataset: (79, 8)\n",
"Shape of the testing dataset: (20, 8)\n"
]
}
],
"source": [
"# splitting the dataset \n",
"X = df.drop(columns=[\"Fertilizer Name\"])\n",
"y = df[\"Fertilizer Name\"]\n",
"\n",
"X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
"\n",
"print(f\"Shape of the training dataset: {X_train.shape}\")\n",
"print(f\"Shape of the testing dataset: {X_test.shape}\")"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['Temparature', 'Humidity ', 'Moisture', 'Nitrogen', 'Potassium', 'Phosphorous']\n"
]
}
],
"source": [
"# numerical columns in the dataset\n",
"print(df._get_numeric_data().columns.tolist())"
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['Soil Type', 'Crop Type']"
]
},
"execution_count": 11,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# segrating categorical columns\n",
"categorical_columns = [i for i in df.columns if (i not in df._get_numeric_data().columns) & (i !='Fertilizer Name')]\n",
"categorical_columns"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Encoding"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
"ohe = OneHotEncoder(drop='first')\n",
"standard_scaler = StandardScaler()\n",
"\n",
"preprocessor = ColumnTransformer(\n",
" transformers =[\n",
" ('StandaradScaling', standard_scaler, df._get_numeric_data().columns),\n",
" ('One_hot_encoding', ohe, categorical_columns)\n",
" ],\n",
" remainder='passthrough'\n",
")\n",
"\n",
"pipeline = Pipeline([\n",
" ('preprocess', preprocessor)\n",
"])"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [],
"source": [
"X_train_transformed = pipeline.fit_transform(X_train)\n",
"X_test_transformed = pipeline.transform(X_test)"
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"le = LabelEncoder()\n",
"\n",
"y_train_transformed = le.fit_transform(y_train)\n",
"y_test_transformed = le.transform(y_test)"
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"def evaluate_clf(true, predicted):\n",
" '''\n",
" This function takes in true values and predicted values\n",
" Returns: Accuracy, F1-Score, Precision, Recall, Roc-auc Score\n",
" '''\n",
" acc = accuracy_score(true, predicted)\n",
" f1 = f1_score(true, predicted, average='weighted')\n",
" precision = precision_score(true, predicted, average='weighted')\n",
" recall = recall_score(true, predicted, average='weighted')\n",
" \n",
" return acc, f1, precision, recall"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"# create a function which can evaluate models and returns a report \n",
"def evaluate_model(X_train, X_test, y_train, y_test, models):\n",
" '''\n",
" This function takes X_train, X_test, y_train, y_test and models dictionary as input\n",
" Iterate through the given model directory and evaluate metrics\n",
"\n",
" Returns:\n",
" DataFrame which contains report of all models metrics \n",
" '''\n",
"\n",
" model_list = []\n",
" metric_list = []\n",
"\n",
" for i in range(len(list(models))):\n",
" model = list(models.values())[i]\n",
" model.fit(X_train, y_train)\n",
"\n",
" # Make predictions\n",
" y_train_pred = model.predict(X_train)\n",
" y_test_pred = model.predict(X_test)\n",
"\n",
" # Training set performances\n",
" model_train_accuracy, model_train_f1, model_train_precision, \\\n",
" model_train_recall = evaluate_clf(y_train, y_train_pred)\n",
"\n",
" # Test set peformances \n",
" model_test_accuracy, model_test_f1, model_test_precision, \\\n",
" model_test_recall = evaluate_clf(y_test, y_test_pred)\n",
"\n",
" print(list(models.keys())[i])\n",
" model_list.append(list(models.keys())[i])\n",
"\n",
" result_dict ={'model_name':list(models.keys())[i], \n",
" \"train_accuracy\": model_train_accuracy, \"test_accuracy\": model_test_accuracy,\n",
" \"train_precision\": model_train_precision, \"test_precision\": model_test_precision,\n",
" 'train_recall': model_train_recall, \"test_recall\":model_test_recall,\n",
" \"train_f1_score\": model_train_f1, \"test_f1_score\": model_test_f1}\n",
"\n",
" metric_list.append(result_dict)\n",
"\n",
" \n",
" return metric_list\n"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"# Model Dictionary\n",
"models = {\n",
" \"Random Forest\": RandomForestClassifier(),\n",
" \"Decision Tree\": DecisionTreeClassifier(),\n",
" \"Gradient Boosting\": GradientBoostingClassifier(),\n",
" \"K-Neighbors Classifier\": KNeighborsClassifier(),\n",
" \"XGBClassifier\": XGBClassifier(), \n",
" \"CatBoosting Classifier\": CatBoostClassifier(verbose=False),\n",
" \"AdaBoost Classifier\": AdaBoostClassifier()\n",
"}"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Random Forest\n",
"Decision Tree\n",
"Gradient Boosting\n",
"K-Neighbors Classifier\n",
"XGBClassifier\n",
"CatBoosting Classifier\n",
"AdaBoost Classifier\n"
]
}
],
"source": [
"resultant_metrics = evaluate_model(X_train_transformed, X_test_transformed, y_train_transformed, y_test_transformed, models)\n",
"\n",
"resultant_metrics_df = pd.DataFrame(data=resultant_metrics)"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" model_name | \n",
" train_accuracy | \n",
" test_accuracy | \n",
" train_precision | \n",
" test_precision | \n",
" train_recall | \n",
" test_recall | \n",
" train_f1_score | \n",
" test_f1_score | \n",
"
\n",
" \n",
" \n",
" \n",
" | 1 | \n",
" Decision Tree | \n",
" 1.000000 | \n",
" 1.00 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.00 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
"
\n",
" \n",
" | 4 | \n",
" XGBClassifier | \n",
" 1.000000 | \n",
" 1.00 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.00 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
"
\n",
" \n",
" | 5 | \n",
" CatBoosting Classifier | \n",
" 1.000000 | \n",
" 1.00 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.00 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
"
\n",
" \n",
" | 0 | \n",
" Random Forest | \n",
" 1.000000 | \n",
" 0.95 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 1.000000 | \n",
" 0.95 | \n",
" 1.000000 | \n",
" 0.966667 | \n",
"
\n",
" \n",
" | 2 | \n",
" Gradient Boosting | \n",
" 1.000000 | \n",
" 0.95 | \n",
" 1.000000 | \n",
" 0.975000 | \n",
" 1.000000 | \n",
" 0.95 | \n",
" 1.000000 | \n",
" 0.955556 | \n",
"
\n",
" \n",
" | 6 | \n",
" AdaBoost Classifier | \n",
" 0.594937 | \n",
" 0.70 | \n",
" 0.477918 | \n",
" 0.657143 | \n",
" 0.594937 | \n",
" 0.70 | \n",
" 0.504147 | \n",
" 0.662500 | \n",
"
\n",
" \n",
" | 3 | \n",
" K-Neighbors Classifier | \n",
" 0.898734 | \n",
" 0.65 | \n",
" 0.904539 | \n",
" 0.666667 | \n",
" 0.898734 | \n",
" 0.65 | \n",
" 0.897599 | \n",
" 0.647727 | \n",
"
\n",
" \n",
"
\n",
"