diff --git a/.idea/.gitignore b/.idea/.gitignore new file mode 100644 index 0000000..b58b603 --- /dev/null +++ b/.idea/.gitignore @@ -0,0 +1,5 @@ +# Default ignored files +/shelf/ +/workspace.xml +# Editor-based HTTP Client requests +/httpRequests/ diff --git a/.idea/Task 2.iml b/.idea/Task 2.iml new file mode 100644 index 0000000..24643cc --- /dev/null +++ b/.idea/Task 2.iml @@ -0,0 +1,12 @@ + + + + + + + + + + + + \ No newline at end of file diff --git a/.idea/modules.xml b/.idea/modules.xml new file mode 100644 index 0000000..6e94f6e --- /dev/null +++ b/.idea/modules.xml @@ -0,0 +1,8 @@ + + + + + + + + \ No newline at end of file diff --git a/About.html b/About.html new file mode 100644 index 0000000..7260744 --- /dev/null +++ b/About.html @@ -0,0 +1,17 @@ + + + + + About me + + +
+

About me

+ 
+        Hello there! My name is Ali,
+        I'm a sophomore student at IAU CCSIT,
+        Also a member of Programming Club(;
+
+
+ + \ No newline at end of file diff --git a/Exercises/day_3/Day3_exercise.ipynb b/Exercises/day_3/Day3_exercise.ipynb new file mode 100644 index 0000000..e6d7ebd --- /dev/null +++ b/Exercises/day_3/Day3_exercise.ipynb @@ -0,0 +1,1607 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "header_image" + }, + "source": [ + "![image.png](https://i.imgur.com/a3uAqnb.png)" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "title_cell" + }, + "source": [ + "# **Neural Networks for Regression: Diamond Price Prediction** 💎\n", + "---" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "intro_cell" + }, + "source": [ + "In this exercise, you will:\n", + "- Predict **diamond prices** using a neural network built with PyTorch\n", + "- Build a **Three-layer neural network regressor**\n", + "- Train the model on tabular diamond data\n", + "- Evaluate the model's performance using **multiple metrics**\n", + "\n", + "---\n", + "\n", + "## **Tasks Overview**\n", + "\n", + "| Task | Description |\n", + "|------|-------------|\n", + "| **Task 1** | Load & Explore Data |\n", + "| **Task 2** | Data Preprocessing (Encoding & Scaling) |\n", + "| **Task 3** | Create PyTorch DataLoaders |\n", + "| **Task 4** | Build the Neural Network Model |\n", + "| **Task 5** | Define Training & Validation Functions |\n", + "| **Task 6** | Train, Evaluate & Visualize Results |" + ] + }, + { + "cell_type": "code", + "execution_count": 32, + "metadata": { + "id": "imports_cell", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "6b0c534d-57c1-4c8b-abd6-c9ccb4dd1dce" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "PyTorch version: 2.9.0+cu126\n", + "CUDA available: True\n" + ] + } + ], + "source": [ + "# Import required libraries\n", + "import kagglehub\n", + "import pandas as pd\n", + "import numpy as np\n", + "import torch\n", + "import torch.nn as nn\n", + "from torch.optim import AdamW\n", + "import matplotlib.pyplot as plt\n", + "from torch.utils.data import DataLoader, TensorDataset\n", + "from sklearn.model_selection import train_test_split\n", + "from sklearn.preprocessing import StandardScaler, LabelEncoder, OneHotEncoder\n", + "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n", + "\n", + "import os\n", + "\n", + "print(f\"PyTorch version: {torch.__version__}\")\n", + "print(f\"CUDA available: {torch.cuda.is_available()}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "about_dataset" + }, + "source": [ + "---\n", + "\n", + "# Task 1: Load & Explore Data\n", + "\n", + "## 📊 **About The Dataset**\n", + "\n", + "This is the classic **Diamonds dataset** used for **regression tasks**. \n", + "It contains features like **carat, cut, color, clarity, depth, table**, and physical dimensions (**x, y, z**).\n", + "\n", + "The **target variable** is **price**, a continuous value we want to predict.\n", + "\n", + "Dataset link: https://www.kaggle.com/datasets/natedir/diamonds/data" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "read_data_header" + }, + "source": [ + "### 🔹**Read Data**" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "id": "download_data", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 241 + }, + "outputId": "a4879486-3301-4fab-8ec5-85ee71b01760" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Using Colab cache for faster access to the 'diamonds' dataset.\n", + "Path to dataset files: /kaggle/input/diamonds\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " carat cut color clarity depth table price x y z\n", + "0 0.23 Ideal E SI2 61.5 55.0 326 3.95 3.98 2.43\n", + "1 0.21 Premium E SI1 59.8 61.0 326 3.89 3.84 2.31\n", + "2 0.23 Good E VS1 56.9 65.0 327 4.05 4.07 2.31\n", + "3 0.29 Premium I VS2 62.4 58.0 334 4.20 4.23 2.63\n", + "4 0.31 Good J SI2 63.3 58.0 335 4.34 4.35 2.75" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
caratcutcolorclaritydepthtablepricexyz
00.23IdealESI261.555.03263.953.982.43
10.21PremiumESI159.861.03263.893.842.31
20.23GoodEVS156.965.03274.054.072.31
30.29PremiumIVS262.458.03344.204.232.63
40.31GoodJSI263.358.03354.344.352.75
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "df", + "summary": "{\n \"name\": \"df\",\n \"rows\": 53940,\n \"fields\": [\n {\n \"column\": \"carat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.47401124440538067,\n \"min\": 0.2,\n \"max\": 5.01,\n \"num_unique_values\": 273,\n \"samples\": [\n 0.77,\n 1.58,\n 0.65\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"cut\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 5,\n \"samples\": [\n \"Premium\",\n \"Fair\",\n \"Good\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"color\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 7,\n \"samples\": [\n \"E\",\n \"I\",\n \"G\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"clarity\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 8,\n \"samples\": [\n \"SI1\",\n \"VVS1\",\n \"SI2\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"depth\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.4326213188337733,\n \"min\": 43.0,\n \"max\": 79.0,\n \"num_unique_values\": 184,\n \"samples\": [\n 61.0,\n 62.9,\n 70.2\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"table\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 2.234490562820938,\n \"min\": 43.0,\n \"max\": 95.0,\n \"num_unique_values\": 127,\n \"samples\": [\n 54.8,\n 61.5,\n 57.6\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"price\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 3989,\n \"min\": 326,\n \"max\": 18823,\n \"num_unique_values\": 11602,\n \"samples\": [\n 16368,\n 895,\n 748\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"x\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1217607467924422,\n \"min\": 0.0,\n \"max\": 10.74,\n \"num_unique_values\": 554,\n \"samples\": [\n 5.88,\n 5.69,\n 4.61\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"y\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1421346741235396,\n \"min\": 0.0,\n \"max\": 58.9,\n \"num_unique_values\": 552,\n \"samples\": [\n 5.33,\n 5.61,\n 5.25\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"z\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.7056988469499964,\n \"min\": 0.0,\n \"max\": 31.8,\n \"num_unique_values\": 375,\n \"samples\": [\n 3.36,\n 2.45,\n 2.7\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 12 + } + ], + "source": [ + "# Download latest version\n", + "path = kagglehub.dataset_download(\"natedir/diamonds\")\n", + "\n", + "print(\"Path to dataset files:\", path)\n", + "df = pd.read_csv(path + \"/diamonds.csv\")\n", + "df = df.drop('Unnamed: 0', axis=1)\n", + "df.head()" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": { + "id": "data_info", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "0bbcc923-3feb-47f0-8930-4061d0a6ef3e" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\n", + "RangeIndex: 53940 entries, 0 to 53939\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 carat 53940 non-null float64\n", + " 1 cut 53940 non-null object \n", + " 2 color 53940 non-null object \n", + " 3 clarity 53940 non-null object \n", + " 4 depth 53940 non-null float64\n", + " 5 table 53940 non-null float64\n", + " 6 price 53940 non-null int64 \n", + " 7 x 53940 non-null float64\n", + " 8 y 53940 non-null float64\n", + " 9 z 53940 non-null float64\n", + "dtypes: float64(6), int64(1), object(3)\n", + "memory usage: 4.1+ MB\n" + ] + } + ], + "source": [ + "df.info()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "prepare_data_header" + }, + "source": [ + "---\n", + "\n", + "# Task 2: Data Preprocessing\n", + "\n", + "### 🔹**Prepare Data**\n", + "\n", + "> Before training the model, we'll prepare the data by **encoding categorical features** and **scaling numerical features** to ensure stable and efficient neural network training.\n", + "\n", + "**Preprocessing Checklist:**\n", + "\n", + "| Step | Action |\n", + "|------|--------|\n", + "| 1 | Encode categorical columns (cut, color, clarity) |\n", + "| 2 | Scale numerical features (but NOT the target!) |\n", + "| 3 | Split into features (X) and target (y) |\n", + "| 4 | Convert to PyTorch tensors |" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "id": "encode_data", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 258 + }, + "outputId": "c7034b4d-b715-4bd4-9d65-72b17c396b9d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Encoding column: cut\n", + "Encoding column: color\n", + "Encoding column: clarity\n" + ] + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " carat cut color clarity depth table price x y z\n", + "0 0.23 2 1 3 61.5 55.0 326 3.95 3.98 2.43\n", + "1 0.21 3 1 2 59.8 61.0 326 3.89 3.84 2.31\n", + "2 0.23 1 1 4 56.9 65.0 327 4.05 4.07 2.31\n", + "3 0.29 3 5 5 62.4 58.0 334 4.20 4.23 2.63\n", + "4 0.31 1 6 3 63.3 58.0 335 4.34 4.35 2.75" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
caratcutcolorclaritydepthtablepricexyz
00.2321361.555.03263.953.982.43
10.2131259.861.03263.893.842.31
20.2311456.965.03274.054.072.31
30.2935562.458.03344.204.232.63
40.3116363.358.03354.344.352.75
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "df", + "summary": "{\n \"name\": \"df\",\n \"rows\": 53940,\n \"fields\": [\n {\n \"column\": \"carat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.47401124440538067,\n \"min\": 0.2,\n \"max\": 5.01,\n \"num_unique_values\": 273,\n \"samples\": [\n 0.77,\n 1.58,\n 0.65\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"cut\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1,\n \"min\": 0,\n \"max\": 4,\n \"num_unique_values\": 5,\n \"samples\": [\n 3,\n 0,\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"color\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1,\n \"min\": 0,\n \"max\": 6,\n \"num_unique_values\": 7,\n \"samples\": [\n 1,\n 5,\n 3\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"clarity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1,\n \"min\": 0,\n \"max\": 7,\n \"num_unique_values\": 8,\n \"samples\": [\n 2,\n 6,\n 3\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"depth\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.4326213188337733,\n \"min\": 43.0,\n \"max\": 79.0,\n \"num_unique_values\": 184,\n \"samples\": [\n 61.0,\n 62.9,\n 70.2\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"table\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 2.234490562820938,\n \"min\": 43.0,\n \"max\": 95.0,\n \"num_unique_values\": 127,\n \"samples\": [\n 54.8,\n 61.5,\n 57.6\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"price\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 3989,\n \"min\": 326,\n \"max\": 18823,\n \"num_unique_values\": 11602,\n \"samples\": [\n 16368,\n 895,\n 748\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"x\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1217607467924422,\n \"min\": 0.0,\n \"max\": 10.74,\n \"num_unique_values\": 554,\n \"samples\": [\n 5.88,\n 5.69,\n 4.61\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"y\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1421346741235396,\n \"min\": 0.0,\n \"max\": 58.9,\n \"num_unique_values\": 552,\n \"samples\": [\n 5.33,\n 5.61,\n 5.25\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"z\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.7056988469499964,\n \"min\": 0.0,\n \"max\": 31.8,\n \"num_unique_values\": 375,\n \"samples\": [\n 3.36,\n 2.45,\n 2.7\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 14 + } + ], + "source": [ + "# TODO: Encode categorical features using LabelEncoder\n", + "# [HINT]: Use select_dtypes to find categorical columns, then loop through them\n", + "\n", + "categorical_cols = df.select_dtypes('object') # \n", + "\n", + "for col in categorical_cols:\n", + " print(f\"Encoding column: {col}\")\n", + " le = LabelEncoder()\n", + " # TODO: Apply fit_transform to encode the column\n", + " df[col] = le.fit_transform(df[col].astype(str)) # \n", + "\n", + "df.head()" + ] + }, + { + "cell_type": "code", + "source": [ + "df.info()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "1Wyi3bqvcKlY", + "outputId": "5b639f26-18d0-4c98-c4ea-6defb87f0221" + }, + "execution_count": 15, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "\n", + "RangeIndex: 53940 entries, 0 to 53939\n", + "Data columns (total 10 columns):\n", + " # Column Non-Null Count Dtype \n", + "--- ------ -------------- ----- \n", + " 0 carat 53940 non-null float64\n", + " 1 cut 53940 non-null int64 \n", + " 2 color 53940 non-null int64 \n", + " 3 clarity 53940 non-null int64 \n", + " 4 depth 53940 non-null float64\n", + " 5 table 53940 non-null float64\n", + " 6 price 53940 non-null int64 \n", + " 7 x 53940 non-null float64\n", + " 8 y 53940 non-null float64\n", + " 9 z 53940 non-null float64\n", + "dtypes: float64(6), int64(4)\n", + "memory usage: 4.1 MB\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": { + "id": "scale_data", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "outputId": "6c919971-e605-4220-f93f-b484e7c362be" + }, + "outputs": [ + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " carat cut color clarity depth table price \\\n", + "0 -1.198168 -0.538099 -0.937163 -0.484264 -0.174092 -1.099672 326 \n", + "1 -1.240361 0.434949 -0.937163 -1.064117 -1.360738 1.585529 326 \n", + "2 -1.198168 -1.511147 -0.937163 0.095589 -3.385019 3.375663 327 \n", + "3 -1.071587 0.434949 1.414272 0.675442 0.454133 0.242928 334 \n", + "4 -1.029394 -1.511147 2.002131 -0.484264 1.082358 0.242928 335 \n", + "\n", + " x y z \n", + "0 -1.587837 -1.536196 -1.571129 \n", + "1 -1.641325 -1.658774 -1.741175 \n", + "2 -1.498691 -1.457395 -1.741175 \n", + "3 -1.364971 -1.317305 -1.287720 \n", + "4 -1.240167 -1.212238 -1.117674 " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
caratcutcolorclaritydepthtablepricexyz
0-1.198168-0.538099-0.937163-0.484264-0.174092-1.099672326-1.587837-1.536196-1.571129
1-1.2403610.434949-0.937163-1.064117-1.3607381.585529326-1.641325-1.658774-1.741175
2-1.198168-1.511147-0.9371630.095589-3.3850193.375663327-1.498691-1.457395-1.741175
3-1.0715870.4349491.4142720.6754420.4541330.242928334-1.364971-1.317305-1.287720
4-1.029394-1.5111472.002131-0.4842641.0823580.242928335-1.240167-1.212238-1.117674
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "df", + "summary": "{\n \"name\": \"df\",\n \"rows\": 53940,\n \"fields\": [\n {\n \"column\": \"carat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696877427,\n \"min\": -1.2614580300362754,\n \"max\": 8.88607463710788,\n \"num_unique_values\": 273,\n \"samples\": [\n -0.05894376386742541,\n 1.6498922985830458,\n -0.3121046620082359\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"cut\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696874767,\n \"min\": -2.4841956038221564,\n \"max\": 1.407997669534476,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.4349493511953178,\n -2.4841956038221564,\n -1.5111472854829984\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"color\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.000009269687875,\n \"min\": -1.525021467723093,\n \"max\": 2.0021308274359626,\n \"num_unique_values\": 7,\n \"samples\": [\n -0.9371627518632505,\n 1.4142721115761203,\n 0.23855467985643486\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"clarity\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696876808,\n \"min\": -2.223823201150214,\n \"max\": 1.8351475642185129,\n \"num_unique_values\": 8,\n \"samples\": [\n -1.0641172681877207,\n 1.255294597737266,\n -0.4842643017064739\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"depth\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696876761,\n \"min\": -13.087602738397585,\n \"max\": 12.04139208771922,\n \"num_unique_values\": 184,\n \"samples\": [\n -0.523105325339182,\n 0.8031471793725374,\n 5.898748908001782\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"table\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696877552,\n \"min\": -6.470073385931448,\n \"max\": 16.801665993491635,\n \"num_unique_values\": 127,\n \"samples\": [\n -1.189178680600826,\n 1.8092954317479188,\n 0.06391497829118802\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"price\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 3989,\n \"min\": 326,\n \"max\": 18823,\n \"num_unique_values\": 11602,\n \"samples\": [\n 16368,\n 895,\n 748\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"x\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696876721,\n \"min\": -5.109120063383867,\n \"max\": 4.465202792140503,\n \"num_unique_values\": 554,\n \"samples\": [\n 0.13268798042277088,\n -0.0366901706526132,\n -0.9994712399232203\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"y\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.0000092696876821,\n \"min\": -5.020930755324094,\n \"max\": 46.54964784548917,\n \"num_unique_values\": 552,\n \"samples\": [\n -0.35418739467324983,\n -0.10903014495800831,\n -0.4242323231633188\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"z\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.000009269687696,\n \"min\": -5.014556274993803,\n \"max\": 40.047575983698835,\n \"num_unique_values\": 375,\n \"samples\": [\n -0.25327437596212776,\n -1.5427882236165393,\n -1.188526177557635\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 24 + } + ], + "source": [ + "# TODO: Standardize features using StandardScaler (DON'T scale the target!)\n", + "# [HINT]: Select numerical columns, but drop \"price\" from the list\n", + "\n", + "numerical_cols = df.select_dtypes(['float', 'int']).columns.drop('price') # \n", + "\n", + "scaler = StandardScaler()\n", + "\n", + "# TODO: Apply fit_transform to scale the numerical columns\n", + "df[numerical_cols] = scaler.fit_transform(df[numerical_cols]) # \n", + "\n", + "df.head()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "split_note" + }, + "source": [ + "> After preprocessing data, we'll first split the dataset into features and target labels, apply a train–test split, and then convert the data into PyTorch tensors" + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "metadata": { + "id": "split_data" + }, + "outputs": [], + "source": [ + "# TODO: Split features from target\n", + "X = df.drop('price', axis=1) # \n", + "y = df['price'] # \n", + "\n", + "# TODO: Split the dataset into training and testing sets (80% train, 20% test)\n", + "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42) # \n", + "\n", + "# TODO: Transform to PyTorch tensors with dtype=torch.float32\n", + "X_train = torch.tensor(X_train.values, dtype=torch.float32) # \n", + "X_test = torch.tensor(X_test.values, dtype=torch.float32) # \n", + "y_train = torch.tensor(y_train.values, dtype=torch.float32) # \n", + "y_test = torch.tensor(y_test.values, dtype=torch.float32) # " + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "dataloader_note" + }, + "source": [ + "---\n", + "\n", + "# Task 3: Create PyTorch DataLoaders\n", + "\n", + "> After converting the data into tensors, we group features and labels into **TensorDataset** objects. \n", + "We then use **DataLoaders** to load the data in mini-batches for efficient training and evaluation.\n", + "\n", + "**Why use DataLoaders?**\n", + "- Handles **batching** automatically\n", + "- Enables **shuffling** of training data\n", + "- Makes training loops cleaner and more efficient" + ] + }, + { + "cell_type": "code", + "execution_count": 38, + "metadata": { + "id": "create_dataloaders", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "ec31ef8a-a172-4112-92ce-712b3b5596b9" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Number of available CPU cores: 2\n", + "Train dataset: 43152\n", + "Test dataset: 10788\n", + "Training batch input shape: torch.Size([32, 9])\n", + "Training batch labels shape: torch.Size([32])\n" + ] + } + ], + "source": [ + "# TODO: Create TensorDatasets for training and testing\n", + "# [HINT]: Use TensorDataset(features, labels)\n", + "train_dataset = TensorDataset(X_train, y_train) # \n", + "test_dataset = TensorDataset(X_test, y_test) # \n", + "\n", + "num_workers = os.cpu_count()\n", + "print(f\"Number of available CPU cores: {num_workers}\")\n", + "\n", + "# TODO: Create DataLoaders with batch_size=32\n", + "# [HINT]: shuffle=True for training, shuffle=False for testing\n", + "train_loader = DataLoader(train_dataset, batch_size=32, shuffle=True, num_workers=num_workers) # \n", + "test_loader = DataLoader(test_dataset, batch_size=32, shuffle=True, num_workers=num_workers) # \n", + "\n", + "# Print dataset sizes\n", + "print(\"Train dataset:\", len(train_dataset))\n", + "print(\"Test dataset:\", len(test_dataset))\n", + "\n", + "# Get the first batch from the training DataLoader\n", + "X_batch, y_batch = next(iter(train_loader))\n", + "print(f\"Training batch input shape: {X_batch.shape}\")\n", + "print(f\"Training batch labels shape: {y_batch.shape}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "data_ready" + }, + "source": [ + "> Data is now ready for the model! Let's build the model class.\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "model_header" + }, + "source": [ + "---\n", + "\n", + "# Task 4: Build the Neural Network Model\n", + "\n", + "### 🔹**Model Class**\n", + "\n", + "Let's create the **architecture of our model** by implementing a **three-layer neural network regressor**.\n", + "\n", + "For regression, the key differences from classification are:\n", + "- **Output layer** produces a single continuous value (not class probabilities)\n", + "- **No softmax** is needed at the output\n", + "- We use **MSE (Mean Squared Error)** loss instead of Cross-Entropy\n", + "\n", + "**Architecture Overview:**\n", + "```\n", + "Input (9 features) → Linear → ReLU → Linear → ReLU → Linear → Output (1 value)\n", + "```" + ] + }, + { + "cell_type": "code", + "execution_count": 55, + "metadata": { + "id": "model_class" + }, + "outputs": [], + "source": [ + "class NN3Layer(nn.Module):\n", + " def __init__(self, input_dim, hidden_dim):\n", + " super(NN3Layer, self).__init__()\n", + " # TODO: Define the first linear layer: input_dim -> hidden_dim\n", + " self.layer1 = nn.Linear(input_dim, hidden_dim) # \n", + "\n", + " # TODO: Define the second linear layer: hidden_dim -> hidden_dim\n", + " self.layer2 = nn.Linear(hidden_dim, hidden_dim) # \n", + "\n", + " # TODO: Define the output layer: hidden_dim -> 1 (single value for regression)\n", + " self.layer3 = nn.Linear(hidden_dim, 1) # \n", + "\n", + " # TODO: Define ReLU activation\n", + " self.relu = nn.ReLU() # \n", + "\n", + " def forward(self, x):\n", + " # TODO: First hidden layer with ReLU\n", + " a1 = self.relu(self.layer1(x)) # \n", + "\n", + " # TODO: Second hidden layer with ReLU\n", + " a2 = self.relu(self.layer2(a1)) # \n", + "\n", + " # TODO: Output layer (no activation for regression)\n", + " output = self.layer3(a2) # \n", + "\n", + " return output" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "training_header" + }, + "source": [ + "---\n", + "\n", + "# Task 5: Define Training & Validation Functions\n", + "\n", + "### 🔹**Training Loop**\n", + "\n", + "> The training loop performs **forward pass**, computes **loss**, does **backpropagation**, and **updates weights**.\n", + "\n", + "**Training Steps per Batch:**\n", + "1. Move data to device (GPU)\n", + "2. Forward pass: Get model predictions\n", + "3. Compute loss (MSE for regression)\n", + "4. Zero gradients, Backward pass, Update weights" + ] + }, + { + "cell_type": "code", + "execution_count": 56, + "metadata": { + "id": "training_loop" + }, + "outputs": [], + "source": [ + "def train_one_epoch(model, optimizer, criterion, train_loader, device):\n", + " # TODO: Set the model to training mode\n", + " # \n", + " model.train()\n", + "\n", + " running_loss = 0.0\n", + "\n", + " for X_batch, y_batch in train_loader:\n", + " # TODO: Move batch to the selected device\n", + " X_batch = X_batch.to(device) # \n", + " y_batch = y_batch.view(-1, 1).to(device) # [HINT: reshape to (-1, 1)]\n", + "\n", + " # TODO: Forward pass - get model predictions\n", + " outputs = model(X_batch) # \n", + "\n", + " # TODO: Compute loss using criterion\n", + " loss = criterion(outputs, y_batch) # \n", + "\n", + " # TODO: Backward pass & optimization\n", + " # Step 1: Clear previous gradients\n", + " # \n", + " optimizer.zero_grad()\n", + "\n", + " # Step 2: Compute gradients (backward pass)\n", + " # \n", + " loss.backward()\n", + "\n", + " # Step 3: Update model parameters\n", + " # \n", + " optimizer.step()\n", + "\n", + " running_loss += loss.item()\n", + "\n", + " # Calculate average loss over all batches\n", + " avg_loss = running_loss / len(train_loader)\n", + "\n", + " return avg_loss" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "validation_header" + }, + "source": [ + "### 🔹**Validation Loop**\n", + "\n", + "> The validation loop is similar to training, but with key differences:\n", + "> - Model is set to **evaluation mode** (`model.eval()`)\n", + "> - We use `torch.no_grad()` to **disable gradient computation**\n", + "> - No weight updates happen, we just evaluate performance" + ] + }, + { + "cell_type": "code", + "execution_count": 57, + "metadata": { + "id": "validation_loop" + }, + "outputs": [], + "source": [ + "def validate(model, criterion, test_loader, device):\n", + " # TODO: Set the model to evaluation mode\n", + " # \n", + " model.eval()\n", + "\n", + " running_loss = 0.0\n", + "\n", + " # TODO: Disable gradient computation using torch.no_grad()\n", + " with torch.no_grad():\n", + " for X_batch, y_batch in test_loader:\n", + " # TODO: Move data to device\n", + " X_batch = X_batch.to(device) # \n", + " y_batch = y_batch.view(-1, 1).to(device) # [HINT: reshape to (-1, 1)]\n", + "\n", + " # TODO: Forward pass - get model predictions\n", + " outputs = model(X_batch) # \n", + "\n", + " # TODO: Compute loss using criterion\n", + " loss = criterion(outputs, y_batch) # \n", + "\n", + " running_loss += loss.item()\n", + "\n", + " avg_loss = running_loss / len(test_loader)\n", + "\n", + " return avg_loss" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "running_training_header" + }, + "source": [ + "---\n", + "\n", + "# Task 6: Train and Evaluate the Model\n", + "\n", + "### 🔹**Model Setup**\n", + "\n", + "> Now let's put everything together! We'll:\n", + "> 1. Set up the device (GPU if available)\n", + "> 2. Instantiate our model\n", + "> 3. Define the loss function and optimizer\n", + "> 4. Run the training loop" + ] + }, + { + "cell_type": "code", + "source": [ + "# TODO: Set up the device (use GPU if available, otherwise CPU)\n", + "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') # \n", + "print(f\"Using device: {device}\")" + ], + "metadata": { + "id": "oLaCZCuR1NaR", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "4b5f6990-166b-436b-fb52-0fb3d9508e2a" + }, + "execution_count": 58, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Using device: cuda\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": 59, + "metadata": { + "id": "model_setup", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "fa1ba802-1675-4db2-bcc3-2f335aec0894" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model Architecture:\n", + "\n", + "NN3Layer(\n", + " (layer1): Linear(in_features=9, out_features=64, bias=True)\n", + " (layer2): Linear(in_features=64, out_features=64, bias=True)\n", + " (layer3): Linear(in_features=64, out_features=1, bias=True)\n", + " (relu): ReLU()\n", + ")\n", + "\n", + "Total trainable parameters: 4865\n" + ] + } + ], + "source": [ + "# Model parameters\n", + "input_dim = X_train.shape[1] # Number of tabular features\n", + "hidden_dim = 64 # Design choice (feel free to experiment!)\n", + "\n", + "# TODO: Instantiate the model and move it to the device\n", + "model = NN3Layer(input_dim, hidden_dim).to(device) # \n", + "\n", + "# Print the model architecture\n", + "print(\"Model Architecture:\\n\")\n", + "print(model)\n", + "\n", + "# Calculate the total number of trainable parameters\n", + "total_params = sum(p.numel() for p in model.parameters() if p.requires_grad)\n", + "print(f\"\\nTotal trainable parameters: {total_params}\")" + ] + }, + { + "cell_type": "code", + "execution_count": 60, + "metadata": { + "id": "training_setup" + }, + "outputs": [], + "source": [ + "# Hyperparameters (feel free to experiment!)\n", + "num_epochs = 20\n", + "learning_rate = 0.001" + ] + }, + { + "cell_type": "code", + "source": [ + "# TODO: Define criterion (loss function) - use MSELoss for regression\n", + "criterion = nn.MSELoss() # \n", + "\n", + "# TODO: Define optimizer - use AdamW with the model parameters and learning rate\n", + "optimizer = AdamW(model.parameters(), learning_rate) # " + ], + "metadata": { + "id": "8ciNmhde1jXd" + }, + "execution_count": 61, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vF-PCwEH3UCz" + }, + "source": [ + "### 🔹**Run Training**\n", + "\n", + "> Now we're ready to train! The training loop will:\n", + "> - Train for the specified number of epochs\n", + "> - Track both training and validation loss\n", + "> - Print progress after each epoch" + ] + }, + { + "cell_type": "code", + "execution_count": 62, + "metadata": { + "id": "run_training", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "c4f9b65c-c658-4b10-eba6-3176ec467b0b" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Starting Training...\n", + "Epoch [5/20], Train Loss: 1501516.6109, Val Loss: 1296255.0035\n", + "Epoch [10/20], Train Loss: 1362245.9938, Val Loss: 1234571.6919\n", + "Epoch [15/20], Train Loss: 1292260.5992, Val Loss: 1195820.5761\n", + "Epoch [20/20], Train Loss: 1259833.5564, Val Loss: 1184814.5056\n", + "Training Complete!\n" + ] + } + ], + "source": [ + "# Run Training\n", + "train_losses = []\n", + "val_losses = []\n", + "\n", + "print('Starting Training...')\n", + "for epoch in range(num_epochs):\n", + " # TODO: Train one epoch using the train_one_epoch function\n", + " train_loss = train_one_epoch(model, optimizer, criterion, train_loader, device) # \n", + "\n", + " # TODO: Validate using the validate function\n", + " val_loss = validate(model, criterion, test_loader, device) # \n", + "\n", + " train_losses.append(train_loss)\n", + " val_losses.append(val_loss)\n", + "\n", + " if (epoch + 1) % 5 == 0:\n", + " print(\n", + " f'Epoch [{epoch+1}/{num_epochs}], '\n", + " f'Train Loss: {train_loss:.4f}, '\n", + " f'Val Loss: {val_loss:.4f}'\n", + " )\n", + "\n", + "print('Training Complete!')" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "plot_note" + }, + "source": [ + "### 🔹**Visualize Results**\n", + "\n", + "> Let's plot the training and validation losses to see how our model learned over time." + ] + }, + { + "cell_type": "code", + "execution_count": 63, + "metadata": { + "id": "plot_losses", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 507 + }, + "outputId": "91758d92-65cc-4824-a345-b13a6789a2f2" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "# Plotting results\n", + "plt.figure(figsize=(7, 5))\n", + "\n", + "plt.plot(train_losses, label='Train Loss')\n", + "plt.plot(val_losses, label='Validation Loss')\n", + "plt.title('Loss over Epochs')\n", + "plt.xlabel('Epoch')\n", + "plt.ylabel('Loss (MSE)')\n", + "plt.legend()\n", + "\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "4FUb4Kqf3UCz" + }, + "source": [ + "### 📊 **Evaluate with More Metrics**\n", + "\n", + "> Beyond MSE loss, we can evaluate our regression model using additional metrics:\n", + "> - **MAE (Mean Absolute Error)**: Average absolute difference between predictions and actual values\n", + "> - **RMSE (Root Mean Squared Error)**: Square root of MSE, in the same units as the target\n", + "> - **R² Score**: Proportion of variance explained by the model (1.0 = perfect)" + ] + }, + { + "cell_type": "code", + "execution_count": 64, + "metadata": { + "id": "VGIWupJ33UCz", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "95adafc6-fd6b-41a9-bf18-b2b5640aa5f1" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Model Evaluation\n", + " MAE: $586.99\n", + " MSE: 1185578.38\n", + " RMSE: $1088.84\n", + " R2: 0.9254\n" + ] + } + ], + "source": [ + "# Get predictions on test set\n", + "model.eval()\n", + "with torch.no_grad():\n", + " # TODO: Get predictions for X_test\n", + " # [HINT]: Move X_test to device, pass through model, then move back to CPU\n", + " predictions = model(X_test.to(device)).cpu().numpy() # \n", + "\n", + "# TODO: Calculate evaluation metrics\n", + "# [HINT]: Use sklearn metrics - mean_absolute_error, mean_squared_error, r2_score\n", + "mae = mean_absolute_error(y_test.numpy(), predictions) # \n", + "mse = mean_squared_error(y_test.numpy(), predictions) # \n", + "rmse = np.sqrt(mse) # [HINT: use np.sqrt()]\n", + "r2 = r2_score(y_test.numpy(), predictions) # \n", + "\n", + "print(\"Model Evaluation\")\n", + "print(f\" MAE: ${mae:.2f}\")\n", + "print(f\" MSE: {mse:.2f}\")\n", + "print(f\" RMSE: ${rmse:.2f}\")\n", + "print(f\" R2: {r2:.4f}\")" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "vEiCMhqK3UCz" + }, + "source": [ + "### 📈 **Visualize: Predicted vs Actual**\n", + "\n", + "> Let's visualize how well our model's predictions match the actual diamond prices.\n", + "> - A **perfect model** would have all points on the diagonal line\n", + "> - Points **above the line** = model underestimated the price\n", + "> - Points **below the line** = model overestimated the price" + ] + }, + { + "cell_type": "code", + "execution_count": 65, + "metadata": { + "id": "zgSPEJyI3UCz", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 607 + }, + "outputId": "74206b8d-aff6-405f-dfee-13a424a6aab6" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "# Scatter plot: Predicted vs Actual\n", + "plt.figure(figsize=(8, 6))\n", + "\n", + "# Plot: Predicted vs Actual scatter\n", + "plt.scatter(y_test.numpy(), predictions.flatten(), alpha=0.5, s=10, c='steelblue')\n", + "\n", + "# Add perfect prediction line\n", + "min_val = min(y_test.min(), predictions.min())\n", + "max_val = max(y_test.max(), predictions.max())\n", + "plt.plot([min_val, max_val], [min_val, max_val], 'r--', lw=2, label='Perfect Prediction')\n", + "\n", + "plt.xlabel('Actual Price ($)', fontsize=12)\n", + "plt.ylabel('Predicted Price ($)', fontsize=12)\n", + "plt.title('Predicted vs Actual Diamond Prices', fontsize=14)\n", + "plt.legend()\n", + "plt.grid(True, alpha=0.3)\n", + "\n", + "plt.tight_layout()\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "f6FOY6vf3UDA" + }, + "source": [ + "### **Contribution: Yara Alzahrani & Sattam Altwaim** :)\n", + "\n", + "![image.jpeg](https://i.imgur.com/Q4egy5l.jpeg)\n" + ] + } + ], + "metadata": { + "accelerator": "GPU", + "colab": { + "gpuType": "T4", + "provenance": [], + "include_colab_link": true + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + }, + "language_info": { + "name": "python" + } + }, + "nbformat": 4, + "nbformat_minor": 0 +} \ No newline at end of file diff --git a/Main.html b/Main.html new file mode 100644 index 0000000..3981ab3 --- /dev/null +++ b/Main.html @@ -0,0 +1,33 @@ + + + + + Ali Albaqqal - Personal Portfolio + + + + + +
+ 
cout >> "Welcome to my page!"
+
+

I'm Ali Albaqqal

+
+    
IAU CCSIT Student

+
+
+ + +

About me

+ + + +

My portfolio

+ + + + \ No newline at end of file diff --git a/Portfolio.html b/Portfolio.html new file mode 100644 index 0000000..68078c9 --- /dev/null +++ b/Portfolio.html @@ -0,0 +1,32 @@ + + + + + My Portfolio + + +
+

Portfolio

+ 
+        I don't know what to put in this section but..
+        As a conclusion of my career:
+        

+            
First Job

+            
I worked as a cashier in a restaurant for almost 2 months while in high-school (summer job).

+            
After graduating

+            
It was adults life at this moment and I had to make hard decisions.
+                Therefore, I decided to work to improve my financial situation,
+                and I did work for 2 years at King Fahd International Airport.

+            
I want to study

+            
After working for 2 years, having my own car and have more experience.
+            I wanted to study and I started studying at IAU in 2021.

+        

+
+
+ + My GitHub page + + + \ No newline at end of file diff --git a/images/Favicon.jpg b/images/Favicon.jpg new file mode 100644 index 0000000..858e40b Binary files /dev/null and b/images/Favicon.jpg differ diff --git a/images/github - icon.png b/images/github - icon.png new file mode 100644 index 0000000..cb99866 Binary files /dev/null and b/images/github - icon.png differ