mirror of
https://github.com/rasbt/LLMs-from-scratch.git
synced 2025-08-08 08:42:56 +00:00
278 lines
7.9 KiB
Plaintext
278 lines
7.9 KiB
Plaintext
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "8968a681-2db1-4840-bb73-7d6c95986825",
|
|
"metadata": {},
|
|
"source": [
|
|
"<table style=\"width:100%\">\n",
|
|
"<tr>\n",
|
|
"<td style=\"vertical-align:middle; text-align:left;\">\n",
|
|
"<font size=\"2\">\n",
|
|
"Supplementary code for the <a href=\"http://mng.bz/orYv\">Build a Large Language Model From Scratch</a> book by <a href=\"https://sebastianraschka.com\">Sebastian Raschka</a><br>\n",
|
|
"<br>Code repository: <a href=\"https://github.com/rasbt/LLMs-from-scratch\">https://github.com/rasbt/LLMs-from-scratch</a>\n",
|
|
"</font>\n",
|
|
"</td>\n",
|
|
"<td style=\"vertical-align:middle; text-align:left;\">\n",
|
|
"<a href=\"http://mng.bz/orYv\"><img src=\"https://sebastianraschka.com/images/LLMs-from-scratch-images/cover-small.webp\" width=\"100px\"></a>\n",
|
|
"</td>\n",
|
|
"</tr>\n",
|
|
"</table>"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "8b6e1cdd-b14e-4368-bdbb-9bf7ab821791",
|
|
"metadata": {},
|
|
"source": [
|
|
"# Scikit-learn Logistic Regression Model"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"id": "c2a72242-6197-4bef-aa05-696a152350d5",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"100% | 80.23 MB | 4.37 MB/s | 18.38 sec elapsed"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"!python download-prepare-dataset.py"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 14,
|
|
"id": "69f32433-e19c-4066-b806-8f30b408107f",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"import pandas as pd\n",
|
|
"\n",
|
|
"train_df = pd.read_csv(\"train.csv\")\n",
|
|
"val_df = pd.read_csv(\"validation.csv\")\n",
|
|
"test_df = pd.read_csv(\"test.csv\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 16,
|
|
"id": "0808b212-fe91-48d9-80b8-55519f8835d5",
|
|
"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>text</th>\n",
|
|
" <th>label</th>\n",
|
|
" </tr>\n",
|
|
" </thead>\n",
|
|
" <tbody>\n",
|
|
" <tr>\n",
|
|
" <th>0</th>\n",
|
|
" <td>The only reason I saw \"Shakedown\" was that it ...</td>\n",
|
|
" <td>0</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>1</th>\n",
|
|
" <td>This is absolute drivel, designed to shock and...</td>\n",
|
|
" <td>0</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>2</th>\n",
|
|
" <td>Lots of scenes and dialogue are flat-out goofy...</td>\n",
|
|
" <td>1</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>3</th>\n",
|
|
" <td>** and 1/2 stars out of **** Lifeforce is one ...</td>\n",
|
|
" <td>1</td>\n",
|
|
" </tr>\n",
|
|
" <tr>\n",
|
|
" <th>4</th>\n",
|
|
" <td>I learned a thing: you have to take this film ...</td>\n",
|
|
" <td>1</td>\n",
|
|
" </tr>\n",
|
|
" </tbody>\n",
|
|
"</table>\n",
|
|
"</div>"
|
|
],
|
|
"text/plain": [
|
|
" text label\n",
|
|
"0 The only reason I saw \"Shakedown\" was that it ... 0\n",
|
|
"1 This is absolute drivel, designed to shock and... 0\n",
|
|
"2 Lots of scenes and dialogue are flat-out goofy... 1\n",
|
|
"3 ** and 1/2 stars out of **** Lifeforce is one ... 1\n",
|
|
"4 I learned a thing: you have to take this film ... 1"
|
|
]
|
|
},
|
|
"execution_count": 16,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"train_df.head()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "markdown",
|
|
"id": "fae87bc1-14ca-4f89-8e12-49f77b0ec00d",
|
|
"metadata": {},
|
|
"source": [
|
|
"## Scikit-learn baseline"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 17,
|
|
"id": "180318b7-de18-4b05-b84a-ba97c72b9d8e",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"from sklearn.feature_extraction.text import CountVectorizer\n",
|
|
"from sklearn.linear_model import LogisticRegression\n",
|
|
"from sklearn.metrics import accuracy_score"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 20,
|
|
"id": "25090b7c-f516-4be2-8083-3a7187fe4635",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"vectorizer = CountVectorizer()\n",
|
|
"\n",
|
|
"X_train = vectorizer.fit_transform(train_df[\"text\"])\n",
|
|
"X_val = vectorizer.transform(val_df[\"text\"])\n",
|
|
"X_test = vectorizer.transform(test_df[\"text\"])\n",
|
|
"\n",
|
|
"y_train, y_val, y_test = train_df[\"label\"], val_df[\"label\"], test_df[\"label\"]"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 22,
|
|
"id": "0247de3a-88f0-4b9c-becd-157baf3acf49",
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"def eval(model, X_train, y_train, X_val, y_val, X_test, y_test):\n",
|
|
" # Making predictions\n",
|
|
" y_pred_train = model.predict(X_train)\n",
|
|
" y_pred_val = model.predict(X_val)\n",
|
|
" y_pred_test = model.predict(X_test)\n",
|
|
" \n",
|
|
" # Calculating accuracy and balanced accuracy\n",
|
|
" accuracy_train = accuracy_score(y_train, y_pred_train)\n",
|
|
" balanced_accuracy_train = balanced_accuracy_score(y_train, y_pred_train)\n",
|
|
" \n",
|
|
" accuracy_val = accuracy_score(y_val, y_pred_val)\n",
|
|
" balanced_accuracy_val = balanced_accuracy_score(y_val, y_pred_val)\n",
|
|
"\n",
|
|
" accuracy_test = accuracy_score(y_test, y_pred_test)\n",
|
|
" balanced_accuracy_test = balanced_accuracy_score(y_test, y_pred_test)\n",
|
|
" \n",
|
|
" # Printing the results\n",
|
|
" print(f\"Training Accuracy: {accuracy_train*100:.2f}%\")\n",
|
|
" print(f\"Validation Accuracy: {accuracy_val*100:.2f}%\")\n",
|
|
" print(f\"Test Accuracy: {accuracy_test*100:.2f}%\")"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 23,
|
|
"id": "c29c6dfc-f72d-40ab-8cb5-783aad1a15ab",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Training Accuracy: 50.01%\n",
|
|
"Validation Accuracy: 50.14%\n",
|
|
"Test Accuracy: 49.91%\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"from sklearn.dummy import DummyClassifier\n",
|
|
"\n",
|
|
"# Create a dummy classifier with the strategy to predict the most frequent class\n",
|
|
"dummy_clf = DummyClassifier(strategy=\"most_frequent\")\n",
|
|
"dummy_clf.fit(X_train, y_train)\n",
|
|
"\n",
|
|
"eval(dummy_clf, X_train, y_train, X_val, y_val, X_test, y_test)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 24,
|
|
"id": "088a8a3a-3b74-4d10-a51b-cb662569ae39",
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Training Accuracy: 99.80%\n",
|
|
"Validation Accuracy: 88.62%\n",
|
|
"Test Accuracy: 88.85%\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"model = LogisticRegression(max_iter=1000)\n",
|
|
"model.fit(X_train, y_train)\n",
|
|
"eval(model, X_train, y_train, X_val, y_val, X_test, y_test)"
|
|
]
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3 (ipykernel)",
|
|
"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": "3.11.4"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 5
|
|
}
|