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

graphic for ai/ml/dl/ds

Browse Source
pull/34/head
Jen Looper 3 years ago
parent ee0d798a7a
commit 21d1d77b69
5 changed files with 374 additions and 63 deletions
Show Stats Download Patch File Download Diff File

4
1-Introduction/1-intro-to-ML/README.md
Unescape View File

@ -42,7 +42,9 @@ The [human brain](https://www.livescience.com/29365-human-brain.html) perceives
In this curriculum, we are going to cover only the core concepts of Machine Learning that a beginner must know. We cover what we call 'Classical Machine Learning', avoiding the complexities of neural networks, 'Deep Learning' - many-layered model-building - and AI, which we will discuss in a different curriculum. But to understand broader concepts of Artificial Intelligence or Deep Learning, a strong fundamental knowledge of Machine Learning is indispensable, and so we would like to offer it here.
TODO: a diagram like this one: https://softwareengineering.stackexchange.com/questions/366996/distinction-between-ai-ml-neural-networks-deep-learning-and-data-mining
![AI, ML, Deep Learning, Data Science](images/ai-ml-ds.png)
> A diagram showing the relationships between AI, ML, Deep Learning, and Data Science. Infographic by [Jen Looper](https://twitter.com/jenlooper) inspired by [this graphic](https://softwareengineering.stackexchange.com/questions/366996/distinction-between-ai-ml-neural-networks-deep-learning-and-data-mining)
The major motivation behind leveraging Machine Learning is to create automated systems that can learn hidden patterns from data to infer intelligent decisions which seem to be loosely inspired by how the human brain learns certain things based on the data it perceives from the outside world.

BIN
1-Introduction/1-intro-to-ML/images/ai-ml-ds.png
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 59 KiB

BIN
4-Classification/4-Applied/solution/model-kn.onnx
View File

Binary file not shown.

101
4-Classification/4-Applied/solution/recipe-detector.ipynb
Unescape View File

@ -14,13 +14,16 @@
},
"orig_nbformat": 2,
"kernelspec": {
"name": "python37364bit8d3b438fb5fc4430a93ac2cb74d693a7",
"name": "python3",
"display_name": "Python 3.7.0 64-bit ('3.7')"
},
"metadata": {
"interpreter": {
"hash": "70b38d7a306a849643e446cd70466270a13445e5987dfa1344ef2b127438fa4d"
}
},
"interpreter": {
"hash": "70b38d7a306a849643e446cd70466270a13445e5987dfa1344ef2b127438fa4d"
}
},
"nbformat": 4,
@ -28,33 +31,25 @@
"cells": [
{
"cell_type": "code",
"execution_count": 6,
"execution_count": 1,
"metadata": {},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Collecting skl2onnx\n",
" Using cached skl2onnx-1.8.0-py2.py3-none-any.whl (230 kB)\n",
"Requirement already satisfied: six in /Users/jenlooper/Library/Python/3.7/lib/python/site-packages (from skl2onnx) (1.12.0)\n",
"Requirement already satisfied: skl2onnx in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (1.8.0)\n",
"Requirement already satisfied: scikit-learn>=0.19 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (0.24.2)\n",
"Requirement already satisfied: scipy>=1.0 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (1.4.1)\n",
"Requirement already satisfied: six in /Users/jenlooper/Library/Python/3.7/lib/python/site-packages (from skl2onnx) (1.12.0)\n",
"Requirement already satisfied: onnx>=1.2.1 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (1.9.0)\n",
"Requirement already satisfied: onnxconverter-common<1.9,>=1.6.1 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (1.8.1)\n",
"Requirement already satisfied: protobuf in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (3.8.0)\n",
"Requirement already satisfied: scikit-learn>=0.19 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (0.24.2)\n",
"Requirement already satisfied: numpy>=1.15 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from skl2onnx) (1.19.2)\n",
"Collecting onnx>=1.2.1\n",
" Downloading onnx-1.9.0-cp37-cp37m-macosx_10_12_x86_64.whl (12.0 MB)\n",
"\u001b[K |████████████████████████████████| 12.0 MB 6.6 MB/s \n",
"\u001b[?25hCollecting onnxconverter-common<1.9,>=1.6.1\n",
" Downloading onnxconverter_common-1.8.1-py2.py3-none-any.whl (77 kB)\n",
"\u001b[K |████████████████████████████████| 77 kB 8.2 MB/s \n",
"\u001b[?25hRequirement already satisfied: setuptools in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from protobuf->skl2onnx) (45.1.0)\n",
"Requirement already satisfied: joblib>=0.11 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from scikit-learn>=0.19->skl2onnx) (0.16.0)\n",
"Requirement already satisfied: threadpoolctl>=2.0.0 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from scikit-learn>=0.19->skl2onnx) (2.1.0)\n",
"Collecting typing-extensions>=3.6.2.1\n",
" Downloading typing_extensions-3.10.0.0-py3-none-any.whl (26 kB)\n",
"Installing collected packages: typing-extensions, onnx, onnxconverter-common, skl2onnx\n",
"Successfully installed onnx-1.9.0 onnxconverter-common-1.8.1 skl2onnx-1.8.0 typing-extensions-3.10.0.0\n",
"Requirement already satisfied: typing-extensions>=3.6.2.1 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from onnx>=1.2.1->skl2onnx) (3.10.0.0)\n",
"Requirement already satisfied: setuptools in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from protobuf->skl2onnx) (45.1.0)\n",
"\u001b[33mWARNING: You are using pip version 20.2.3; however, version 21.1.2 is available.\n",
"You should consider upgrading via the '/Library/Frameworks/Python.framework/Versions/3.7/bin/python3.7 -m pip install --upgrade pip' command.\u001b[0m\n",
"Note: you may need to restart the kernel to use updated packages.\n"
@ -67,24 +62,19 @@
},
{
"cell_type": "code",
"execution_count": 7,
"execution_count": 2,
"metadata": {},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Collecting onnxruntime\n",
" Downloading onnxruntime-1.8.0-cp37-cp37m-macosx_10_12_x86_64.whl (5.0 MB)\n",
"\u001b[K |████████████████████████████████| 5.0 MB 3.1 MB/s \n",
"\u001b[?25hRequirement already satisfied: numpy>=1.16.6 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from onnxruntime) (1.19.2)\n",
"Requirement already satisfied: onnxruntime in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (1.8.0)\n",
"Requirement already satisfied: protobuf in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from onnxruntime) (3.8.0)\n",
"Collecting flatbuffers\n",
" Downloading flatbuffers-2.0-py2.py3-none-any.whl (26 kB)\n",
"Requirement already satisfied: setuptools in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from protobuf->onnxruntime) (45.1.0)\n",
"Requirement already satisfied: flatbuffers in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from onnxruntime) (2.0)\n",
"Requirement already satisfied: numpy>=1.16.6 in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from onnxruntime) (1.19.2)\n",
"Requirement already satisfied: six>=1.9 in /Users/jenlooper/Library/Python/3.7/lib/python/site-packages (from protobuf->onnxruntime) (1.12.0)\n",
"Installing collected packages: flatbuffers, onnxruntime\n",
"Successfully installed flatbuffers-2.0 onnxruntime-1.8.0\n",
"Requirement already satisfied: setuptools in /Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages (from protobuf->onnxruntime) (45.1.0)\n",
"\u001b[33mWARNING: You are using pip version 20.2.3; however, version 21.1.2 is available.\n",
"You should consider upgrading via the '/Library/Frameworks/Python.framework/Versions/3.7/bin/python3.7 -m pip install --upgrade pip' command.\u001b[0m\n",
"Note: you may need to restart the kernel to use updated packages.\n"
@ -97,7 +87,7 @@
},
{
"cell_type": "code",
"execution_count": 8,
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
@ -107,7 +97,7 @@
},
{
"cell_type": "code",
"execution_count": 9,
"execution_count": 4,
"metadata": {},
"outputs": [
{
@ -140,7 +130,7 @@
"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>Unnamed: 0</th>\n <th>cuisine</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>...</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>indian</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>indian</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>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>2</td>\n <td>indian</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>3</td>\n <td>indian</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>4</td>\n <td>indian</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 × 382 columns</p>\n</div>"
},
"metadata": {},
"execution_count": 9
"execution_count": 4
}
],
"source": [
@ -150,7 +140,7 @@
},
{
"cell_type": "code",
"execution_count": 10,
"execution_count": 5,
"metadata": {},
"outputs": [
{
@ -183,7 +173,7 @@
"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>"
},
"metadata": {},
"execution_count": 10
"execution_count": 5
}
],
"source": [
@ -193,7 +183,7 @@
},
{
"cell_type": "code",
"execution_count": 11,
"execution_count": 6,
"metadata": {},
"outputs": [
{
@ -210,7 +200,7 @@
"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>"
},
"metadata": {},
"execution_count": 11
"execution_count": 6
}
],
"source": [
@ -220,15 +210,11 @@
},
{
"cell_type": "code",
"execution_count": 12,
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"from sklearn.model_selection import train_test_split\n",
"from sklearn.naive_bayes import BernoulliNB,GaussianNB\n",
"from sklearn.neighbors import KNeighborsClassifier\n",
"from sklearn.tree import DecisionTreeClassifier\n",
"from sklearn.ensemble import RandomForestClassifier\n",
"from sklearn.svm import SVC\n",
"from sklearn.model_selection import cross_val_score\n",
"from sklearn.metrics import accuracy_score,precision_score,confusion_matrix,classification_report"
@ -236,7 +222,7 @@
},
{
"cell_type": "code",
"execution_count": 13,
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
@ -245,53 +231,44 @@
},
{
"cell_type": "code",
"execution_count": 14,
"execution_count": 9,
"metadata": {},
"outputs": [
{
"output_type": "execute_result",
"data": {
"text/plain": [
"KNeighborsClassifier()"
"SVC(C=10, kernel='linear', probability=True, random_state=0)"
]
},
"metadata": {},
"execution_count": 14
"execution_count": 9
}
],
"source": [
"# 5 types of model fitting\n",
"model_gaussian = GaussianNB()\n",
"model_gaussian.fit(X_train,y_train.values.ravel())\n",
"model_rfst = RandomForestClassifier()\n",
"model_rfst.fit(X_train,y_train.values.ravel())\n",
"model_nba = BernoulliNB(binarize = .5)\n",
"model_nba.fit(X_train,y_train.values.ravel())\n",
"model_dt = DecisionTreeClassifier()\n",
"model_dt.fit(X_train,y_train.values.ravel())\n",
"model_kn = KNeighborsClassifier()\n",
"model_kn.fit(X_train,y_train.values.ravel())\n"
"model = SVC(kernel='linear', C=10, probability=True,random_state=0)\n",
"model.fit(X_train,y_train.values.ravel())\n"
]
},
{
"cell_type": "code",
"execution_count": 15,
"execution_count": 10,
"metadata": {},
"outputs": [],
"source": [
"y_pred = model_kn.predict(X_test)"
"y_pred = model.predict(X_test)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"execution_count": 11,
"metadata": {},
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
" precision recall f1-score support\n\n chinese 0.59 0.74 0.66 239\n indian 0.83 0.80 0.82 237\n japanese 0.68 0.77 0.72 252\n korean 0.93 0.65 0.76 220\n thai 0.77 0.71 0.74 251\n\n accuracy 0.73 1199\n macro avg 0.76 0.73 0.74 1199\nweighted avg 0.76 0.73 0.74 1199\n\n"
" precision recall f1-score support\n\n chinese 0.68 0.69 0.68 249\n indian 0.92 0.88 0.90 238\n japanese 0.77 0.68 0.72 236\n korean 0.84 0.79 0.82 247\n thai 0.73 0.88 0.80 229\n\n accuracy 0.78 1199\n macro avg 0.79 0.79 0.78 1199\nweighted avg 0.79 0.78 0.78 1199\n\n"
]
}
],
@ -301,7 +278,7 @@
},
{
"cell_type": "code",
"execution_count": 17,
"execution_count": 12,
"metadata": {},
"outputs": [],
"source": [
@ -309,9 +286,9 @@
"from skl2onnx.common.data_types import FloatTensorType\n",
"\n",
"initial_type = [('float_input', FloatTensorType([None, 4]))]\n",
"options = {id(model_kn): {'nocl': True, 'zipmap': False}}\n",
"onx = convert_sklearn(model_kn, initial_types=initial_type,options=options)\n",
"with open(\"./model-kn.onnx\", \"wb\") as f:\n",
"options = {id(model): {'nocl': True, 'zipmap': False}}\n",
"onx = convert_sklearn(model, initial_types=initial_type,options=options)\n",
"with open(\"./model.onnx\", \"wb\") as f:\n",
" f.write(onx.SerializeToString())\n",
"\n",
"\n"

332
5-Clustering/2-K-Means/solution/tester.ipynb
View File

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