msb-public
/
ML-For-Beginners
mirror of https://github.com/microsoft/ML-For-Beginners.git
1
0
Code Issues Projects Releases Wiki Activity

classification starter

Browse Source
pull/34/head
Jen Looper 3 years ago
parent 5d2bee276e
commit c79b78edb8
4 changed files with 2204 additions and 300 deletions
Show Stats Download Patch File Download Diff File

18
4-Classification/1-Introduction/README.md
Unescape View File

@ -1,18 +1,26 @@
# Introduction to Classification
In these four lessons, you will discover the 'meat and potatoes' of classic machine learning - Classification. No pun intended - we will walk through using various classification algorithms with a dataset all about the brilliant cuisines of Asia. Hope you're hungry!
Classification is a form of [supervised learning](https://wikipedia.org/wiki/Supervised_learning) that bears a lot in common with Regression techniques. If machine learning is all about assigning names to things via datasets, then classification generally falls into two groups: binary classification and multiclass classfication.
Remember, Linear Regression helped you predict relationships between variables and make accurate predictions on where a new datapoint would fall in relationship to that line. So, you could predict what price a pumpkin would be in September vs. December, for example. Logistic Regression helped you discover binary categories: at this price point, is this pumpkin orange or not-orange?
Classification uses various algorithms to determine other ways of determining a data point's label or class. Let's work with this recipe data to see whether, by observing a group of ingredients, we can determine its cuisine of origin.
[![Introduction to Classification](https://img.youtube.com/vi/eg8DJYwdMyg/0.jpg)](https://youtu.be/eg8DJYwdMyg "Introduction to Classification")
> 🎥 Click the image above for a video: MIT's John Guttag introduces Classification
## [Pre-lecture quiz](link-to-quiz-app)
Describe what we will learn
### Introduction
Describe what will be covered
Before working to clean the data and prepare it for analysis, it's useful to understand several of the algorithms that you will use.
> Notes
Support-vector machines
Naive Bayes
Decision trees
K-nearest neighbor algorithm
### Prerequisite

1521
4-Classification/1-Introduction/solution/data-prep-visual.ipynb
View File

File diff suppressed because one or more lines are too long

563
4-Classification/1-Introduction/solution/intro-classification.ipynb
Unescape View File

@ -0,0 +1,563 @@
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Build Classification Model"
]
},
{
"cell_type": "code",
"execution_count": 58,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.linear_model import LogisticRegression\r\n",
"from sklearn.model_selection import train_test_split, cross_val_score\r\n",
"from sklearn.metrics import accuracy_score,precision_score,confusion_matrix,classification_report, precision_recall_curve\r\n",
"from sklearn.svm import SVC\r\n",
"import pandas as pd\r\n",
"import numpy as np"
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [
{
"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>almond</th>\n",
" <th>angelica</th>\n",
" <th>anise</th>\n",
" <th>anise_seed</th>\n",
" <th>apple</th>\n",
" <th>apple_brandy</th>\n",
" <th>apricot</th>\n",
" <th>armagnac</th>\n",
" <th>artemisia</th>\n",
" <th>artichoke</th>\n",
" <th>...</th>\n",
" <th>whiskey</th>\n",
" <th>white_bread</th>\n",
" <th>white_wine</th>\n",
" <th>whole_grain_wheat_flour</th>\n",
" <th>wine</th>\n",
" <th>wood</th>\n",
" <th>yam</th>\n",
" <th>yeast</th>\n",
" <th>yogurt</th>\n",
" <th>zucchini</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\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>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>0</td>\n",
" <td>...</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>0</td>\n",
" <td>0</td>\n",
" <td>1</td>\n",
" <td>0</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"<p>5 rows × 380 columns</p>\n",
"</div>"
],
"text/plain": [
" almond angelica anise anise_seed apple apple_brandy apricot \\\n",
"0 0 0 0 0 0 0 0 \n",
"1 1 0 0 0 0 0 0 \n",
"2 0 0 0 0 0 0 0 \n",
"3 0 0 0 0 0 0 0 \n",
"4 0 0 0 0 0 0 0 \n",
"\n",
" armagnac artemisia artichoke ... whiskey white_bread white_wine \\\n",
"0 0 0 0 ... 0 0 0 \n",
"1 0 0 0 ... 0 0 0 \n",
"2 0 0 0 ... 0 0 0 \n",
"3 0 0 0 ... 0 0 0 \n",
"4 0 0 0 ... 0 0 0 \n",
"\n",
" whole_grain_wheat_flour wine wood yam yeast yogurt zucchini \n",
"0 0 0 0 0 0 0 0 \n",
"1 0 0 0 0 0 0 0 \n",
"2 0 0 0 0 0 0 0 \n",
"3 0 0 0 0 0 0 0 \n",
"4 0 0 0 0 0 1 0 \n",
"\n",
"[5 rows x 380 columns]"
]
},
"execution_count": 48,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"transformed_feature_df = pd.read_csv(\".data/features_dataset.csv\")\r\n",
"transformed_feature_df= transformed_feature_df.drop(['Unnamed: 0'], axis=1)\r\n",
"transformed_feature_df.head()"
]
},
{
"cell_type": "code",
"execution_count": 49,
"metadata": {},
"outputs": [
{
"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>cuisine</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>indian</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>indian</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>indian</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>indian</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>indian</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" cuisine\n",
"0 indian\n",
"1 indian\n",
"2 indian\n",
"3 indian\n",
"4 indian"
]
},
"execution_count": 49,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"transformed_label_df = pd.read_csv(\".data/labels_dataset.csv\")\r\n",
"transformed_label_df= transformed_label_df.drop(['Unnamed: 0'], axis=1)\r\n",
"transformed_label_df.head()"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {},
"outputs": [],
"source": [
"X_train, X_test, y_train, y_test = train_test_split(transformed_feature_df, transformed_label_df, test_size=0.3)"
]
},
{
"cell_type": "code",
"execution_count": 59,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Accuracy is 0.8031693077564637\n"
]
}
],
"source": [
"lr = LogisticRegression(multi_class='ovr',solver='lbfgs')\r\n",
"model = lr.fit(X_train, np.ravel(y_train))\r\n",
"\r\n",
"accuracy = model.score(X_test, y_test)\r\n",
"print (\"Accuracy is {}\".format(accuracy))"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"ingredients: Index(['corn'], dtype='object')\n",
"cusine: cuisine thai\n",
"Name: 3816, dtype: object\n"
]
}
],
"source": [
"# test an item\r\n",
"print(f'ingredients: {X_test.iloc[20][X_test.iloc[20]!=0].keys()}')\r\n",
"print(f'cusine: {y_test.iloc[20]}')"
]
},
{
"cell_type": "code",
"execution_count": 53,
"metadata": {},
"outputs": [
{
"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>0</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>thai</th>\n",
" <td>0.475724</td>\n",
" </tr>\n",
" <tr>\n",
" <th>chinese</th>\n",
" <td>0.201912</td>\n",
" </tr>\n",
" <tr>\n",
" <th>japanese</th>\n",
" <td>0.152046</td>\n",
" </tr>\n",
" <tr>\n",
" <th>korean</th>\n",
" <td>0.110980</td>\n",
" </tr>\n",
" <tr>\n",
" <th>indian</th>\n",
" <td>0.059338</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" 0\n",
"thai 0.475724\n",
"chinese 0.201912\n",
"japanese 0.152046\n",
"korean 0.110980\n",
"indian 0.059338"
]
},
"execution_count": 53,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#rehsape to 2d array and transpose\r\n",
"test= X_test.iloc[20].values.reshape(-1, 1).T\r\n",
"# predict with score\r\n",
"proba = model.predict_proba(test)\r\n",
"classes = model.classes_\r\n",
"# create df with classes and scores\r\n",
"resultdf = pd.DataFrame(data=proba, columns=classes)\r\n",
"\r\n",
"# create df to show results\r\n",
"topPrediction = resultdf.T.sort_values(by=[0], ascending = [False])\r\n",
"topPrediction.head()"
]
},
{
"cell_type": "code",
"execution_count": 54,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" precision recall f1-score support\n",
"\n",
" chinese 0.77 0.70 0.74 239\n",
" indian 0.88 0.88 0.88 240\n",
" japanese 0.76 0.79 0.77 227\n",
" korean 0.86 0.78 0.82 240\n",
" thai 0.75 0.86 0.80 253\n",
"\n",
" accuracy 0.80 1199\n",
" macro avg 0.81 0.80 0.80 1199\n",
"weighted avg 0.81 0.80 0.80 1199\n",
"\n"
]
}
],
"source": [
"y_pred = model.predict(X_test)\r\n",
"print(classification_report(y_test,y_pred))"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Try different classifiers"
]
},
{
"cell_type": "code",
"execution_count": 60,
"metadata": {},
"outputs": [],
"source": [
"\r\n",
"C = 10\r\n",
"# Create different classifiers.\r\n",
"classifiers = {\r\n",
" 'L1 logistic': LogisticRegression(C=C, penalty='l1',\r\n",
" solver='saga',\r\n",
" multi_class='multinomial',\r\n",
" max_iter=10000),\r\n",
" 'L2 logistic (Multinomial)': LogisticRegression(C=C, penalty='l2',\r\n",
" solver='saga',\r\n",
" multi_class='multinomial',\r\n",
" max_iter=10000),\r\n",
" 'L2 logistic (OvR)': LogisticRegression(C=C, penalty='l2',\r\n",
" solver='saga',\r\n",
" multi_class='ovr',\r\n",
" max_iter=10000),\r\n",
" 'Linear SVC': SVC(kernel='linear', C=C, probability=True,\r\n",
" random_state=0)\r\n",
"}\r\n"
]
},
{
"cell_type": "code",
"execution_count": 61,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Accuracy (train) for L1 logistic: 79.9% \n",
"Accuracy (train) for L2 logistic (Multinomial): 79.7% \n",
"Accuracy (train) for L2 logistic (OvR): 79.8% \n",
"Accuracy (train) for Linear SVC: 77.9% \n"
]
}
],
"source": [
"n_classifiers = len(classifiers)\r\n",
"\r\n",
"for index, (name, classifier) in enumerate(classifiers.items()):\r\n",
" classifier.fit(X_train, np.ravel(y_train))\r\n",
"\r\n",
" y_pred = classifier.predict(X_test)\r\n",
" accuracy = accuracy_score(y_test, y_pred)\r\n",
" print(\"Accuracy (train) for %s: %0.1f%% \" % (name, accuracy * 100))\r\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"interpreter": {
"hash": "dd61f40108e2a19f4ef0d3ebbc6b6eea57ab3c4bc13b15fe6f390d3d86442534"
},
"kernelspec": {
"display_name": "Python 3.8.5 64-bit ('onnxwine': conda)",
"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": "3.8.5"
}
},
"nbformat": 4,
"nbformat_minor": 4
}

402
4-Classification/1-Introduction/solution/notebook.ipynb
View File

File diff suppressed because one or more lines are too long
Write Preview
Loading…
Cancel
Save