![\"Open](\"https://colab.research.google.com/assets/colab-badge.svg\"){kind=icon}
"
]
},
{
"cell_type": "markdown",
"source": [
"# Setup and Config"
],
"metadata": {
"id": "9yzf5HsVhhQG"
}
},
{
"cell_type": "code",
"source": [
"import os\n",
"import cv2\n",
"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 LabelEncoder\n",
"from keras.preprocessing.image import ImageDataGenerator\n",
"import keras\n",
"from keras.callbacks import EarlyStopping,ModelCheckpoint\n",
"import tensorflow as tf\n",
"from sklearn.metrics import confusion_matrix\n",
"from sklearn.metrics import classification_report\n",
"from tqdm import tqdm"
],
"metadata": {
"id": "xpoaMwyThgc4"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "code",
"source": [
"%%writefile kaggle.json\n",
"{\"username\":\"sairamparshi07\",\"key\":\"a07ce4e9f16b95b9b0c358f5b12deec7\"}"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/"
},
"id": "57HLZ-mkjBFR",
"outputId": "9c092a14-84fe-446e-fba7-4b964beb2776"
},
"execution_count": null,
"outputs": [
{
"output_type": "stream",
"name": "stdout",
"text": [
"Writing kaggle.json\n"
]
}
]
},
{
"cell_type": "code",
"source": [
"import os\n",
"\n",
"# Function to download Kaggle dataset\n",
"def download_kaggle_dataset(dataset_path):\n",
" # Ensure Kaggle API is installed\n",
" os.system('pip install -q kaggle')\n",
"\n",
" # Check for kaggle.json file\n",
" if not os.path.isfile('kaggle.json'):\n",
" print(\"kaggle.json file not found. Please upload it.\")\n",
" return\n",
"\n",
" # Set up Kaggle directory and permissions\n",
" os.system('mkdir -p ~/.kaggle')\n",
" os.system('cp kaggle.json ~/.kaggle/')\n",
" os.system('chmod 600 ~/.kaggle/kaggle.json')\n",
"\n",
" # Download and unzip the dataset\n",
" os.system(f'kaggle datasets download -d {dataset_path}')\n",
" zip_file = f'{dataset_path.split(\"/\")[-1]}.zip'\n",
" os.system(f'unzip -q {zip_file}')\n",
"\n",
"# Example usage\n",
"download_kaggle_dataset('ahmedashrafahmed/bone-fracture')\n"
],
"metadata": {
"id": "G5nREckChhGM"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"# Data Loading\n"
],
"metadata": {
"id": "Cbd0ab1qinqP"
}
},
{
"cell_type": "code",
"source": [
"import pandas as pd\n",
"\n",
"# Function to load data from a given dataset path\n",
"def load_data(dataset_path):\n",
" images, labels = [], []\n",
" for subfolder in os.listdir(dataset_path):\n",
" subfolder_path = os.path.join(dataset_path, subfolder)\n",
" if not os.path.isdir(subfolder_path):\n",
" continue\n",
" for image_filename in os.listdir(subfolder_path):\n",
" if image_filename.endswith(\".jpg\"):\n",
" image_path = os.path.join(subfolder_path, image_filename)\n",
" images.append(image_path)\n",
" labels.append(subfolder)\n",
" return pd.DataFrame({'image': images, 'label': labels})\n",
"\n",
"# Load train, test, and validation data\n",
"train_data = load_data('/content/dataset/train')\n",
"test_data = load_data('/content/dataset/test')\n",
"val_data = load_data('/content/dataset/val')\n"
],
"metadata": {
"id": "fsPnMUMOiodJ"
},
"execution_count": null,
"outputs": []
},
{
"cell_type": "markdown",
"source": [
"# Visualization of Dataset\n"
],
"metadata": {
"id": "c1Gre2PnjHgO"
}
},
{
"cell_type": "code",
"source": [
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import cv2\n",
"import numpy as np\n",
"\n",
"# Function to plot data distribution\n",
"def plot_data_distribution(data, title):\n",
" plt.figure(figsize=(15, 5))\n",
" sns.countplot(x=data['label'])\n",
" plt.xlabel(\"Class\")\n",
" plt.ylabel(\"Number of Samples\")\n",
" plt.title(title)\n",
" plt.xticks(rotation=45)\n",
" plt.grid(True)\n",
"\n",
"# Plotting the distribution for train, test, and validation sets\n",
"plot_data_distribution(train_data, 'Train Data Distribution')\n",
"plot_data_distribution(test_data, 'Test Data Distribution')\n",
"plot_data_distribution(val_data, 'Validation Data Distribution')\n",
"\n",
"# Function to display sample images\n",
"def display_sample_images(data, num_samples=10):\n",
" plt.figure(figsize=(50, 50))\n",
" for i, idx in enumerate(np.random.randint(0, len(data), num_samples)):\n",
" plt.subplot(10, 5, i + 1)\n",
" img = cv2.imread(data.iloc[idx]['image'])\n",
" plt.imshow(img)\n",
" plt.axis('off')\n",
" plt.title(data.iloc[idx]['label'], fontsize=25)\n",
"\n",
"# Display sample images from training set\n",
"display_sample_images(train_data, 50)\n"
],
"metadata": {
"colab": {
"base_uri": "https://localhost:8080/",
"height": 1000
},
"id": "cTd90r5KjIcL",
"outputId": "1c559b5b-a70c-4631-9e25-1d530c64f8f6"
},
"execution_count": null,
"outputs": [
{
"output_type": "display_data",
"data": {
"text/plain": [
"| \n", " | loss | \n", "accuracy | \n", "specificity_at_sensitivity | \n", "auc | \n", "val_loss | \n", "val_accuracy | \n", "val_specificity_at_sensitivity | \n", "val_auc | \n", "
|---|---|---|---|---|---|---|---|---|
| 0 | \n", "3.616755 | \n", "0.724921 | \n", "0.821463 | \n", "0.766576 | \n", "22.047369 | \n", "0.584158 | \n", "0.000000 | \n", "0.500000 | \n", "
| 1 | \n", "0.845531 | \n", "0.865267 | \n", "0.949952 | \n", "0.919102 | \n", "13.399078 | \n", "0.584158 | \n", "0.000000 | \n", "0.500000 | \n", "
| 2 | \n", "0.455430 | \n", "0.903832 | \n", "0.973532 | \n", "0.955804 | \n", "6.155796 | \n", "0.618812 | \n", "0.220238 | \n", "0.606247 | \n", "
| 3 | \n", "0.335553 | \n", "0.927752 | \n", "0.984601 | \n", "0.971034 | \n", "0.544004 | \n", "0.913366 | \n", "0.898810 | \n", "0.944751 | \n", "
| 4 | \n", "0.228689 | \n", "0.934098 | \n", "0.990375 | \n", "0.979109 | \n", "0.148531 | \n", "0.962871 | \n", "0.970238 | \n", "0.983555 | \n", "
| 5 | \n", "0.163920 | \n", "0.956309 | \n", "0.990857 | \n", "0.986656 | \n", "0.053871 | \n", "0.982673 | \n", "1.000000 | \n", "0.997427 | \n", "
| 6 | \n", "0.103800 | \n", "0.969978 | \n", "0.995669 | \n", "0.992711 | \n", "0.016854 | \n", "0.997525 | \n", "1.000000 | \n", "0.999886 | \n", "
| 7 | \n", "0.113678 | \n", "0.966073 | \n", "0.996150 | \n", "0.992422 | \n", "0.027444 | \n", "0.982673 | \n", "1.000000 | \n", "0.999672 | \n", "
| 8 | \n", "0.113922 | \n", "0.972419 | \n", "0.994706 | \n", "0.992036 | \n", "0.112278 | \n", "0.982673 | \n", "0.988095 | \n", "0.993316 | \n", "
| 9 | \n", "0.143280 | \n", "0.968758 | \n", "0.991819 | \n", "0.989606 | \n", "0.915044 | \n", "0.881188 | \n", "1.000000 | \n", "0.939707 | \n", "
| \n", " | Actual | \n", "Prediction | \n", "
|---|---|---|
| 0 | \n", "0 | \n", "0 | \n", "
| 1 | \n", "0 | \n", "0 | \n", "
| 2 | \n", "0 | \n", "0 | \n", "
| 3 | \n", "0 | \n", "0 | \n", "
| 4 | \n", "0 | \n", "0 | \n", "
| ... | \n", "... | \n", "... | \n", "
| 394 | \n", "1 | \n", "1 | \n", "
| 395 | \n", "1 | \n", "1 | \n", "
| 396 | \n", "1 | \n", "1 | \n", "
| 397 | \n", "1 | \n", "1 | \n", "
| 398 | \n", "1 | \n", "1 | \n", "
399 rows × 2 columns
\n", "