noshot 12.0.0__py3-none-any.whl → 14.0.0__py3-none-any.whl

noshot/data/DLE FSD BDA/DLE/DLE 2 (tim stan s)/2.2 DNN vs CNN.ipynb

@@ -0,0 +1,123 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "e859c172-a86c-4cdf-93e3-7c24a9dd3292",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import tensorflow as tf\n",
+    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+    "from tensorflow.keras.models import Sequential\n",
+    "from tensorflow.keras.layers import Dense, Flatten, Conv2D, MaxPooling2D, Dropout\n",
+    "from tensorflow.keras.optimizers import Adam\n",
+    "import matplotlib.pyplot as plt\n",
+    "import pandas as pd\n",
+    "\n",
+    "data_dir = \"Bean_Dataset/Bean_Dataset\"\n",
+    "img_height, img_width = 128, 128\n",
+    "batch_size = 32\n",
+    "\n",
+    "datagen = ImageDataGenerator(rescale=1./255, validation_split=0.2)\n",
+    "\n",
+    "train_gen = datagen.flow_from_directory(\n",
+    "    data_dir, target_size=(img_height, img_width),\n",
+    "    batch_size=batch_size, class_mode='categorical', subset='training'\n",
+    ")\n",
+    "\n",
+    "val_gen = datagen.flow_from_directory(\n",
+    "    data_dir, target_size=(img_height, img_width),\n",
+    "    batch_size=batch_size, class_mode='categorical', subset='validation'\n",
+    ")\n",
+    "\n",
+    "num_classes = len(train_gen.class_indices)\n",
+    "\n",
+    "dnn_model = Sequential([\n",
+    "    Flatten(input_shape=(img_height, img_width, 3)),\n",
+    "    Dense(512, activation='relu'),\n",
+    "    Dropout(0.5),\n",
+    "    Dense(256, activation='relu'),\n",
+    "    Dropout(0.5),\n",
+    "    Dense(num_classes, activation='softmax')\n",
+    "])\n",
+    "dnn_model.compile(optimizer=Adam(), loss='categorical_crossentropy', metrics=['accuracy'])\n",
+    "\n",
+    "cnn_model = Sequential([\n",
+    "    Conv2D(32, (3,3), activation='relu', input_shape=(img_height, img_width, 3)),\n",
+    "    MaxPooling2D(2,2),\n",
+    "    Conv2D(64, (3,3), activation='relu'),\n",
+    "    MaxPooling2D(2,2),\n",
+    "    Conv2D(128, (3,3), activation='relu'),\n",
+    "    MaxPooling2D(2,2),\n",
+    "    Flatten(),\n",
+    "    Dense(512, activation='relu'),\n",
+    "    Dropout(0.5),\n",
+    "    Dense(num_classes, activation='softmax')\n",
+    "])\n",
+    "cnn_model.compile(optimizer=Adam(), loss='categorical_crossentropy', metrics=['accuracy'])\n",
+    "\n",
+    "dnn_history = dnn_model.fit(train_gen, epochs=20, validation_data=val_gen)\n",
+    "cnn_history = cnn_model.fit(train_gen, epochs=20, validation_data=val_gen)\n",
+    "\n",
+    "comparison = pd.DataFrame({\n",
+    "    'Model': ['DNN', 'CNN'],\n",
+    "    'Parameters': [dnn_model.count_params(), cnn_model.count_params()],\n",
+    "    'Final Train Accuracy': [\n",
+    "        dnn_history.history['accuracy'][-1],\n",
+    "        cnn_history.history['accuracy'][-1]\n",
+    "    ],\n",
+    "    'Final Val Accuracy': [\n",
+    "        dnn_history.history['val_accuracy'][-1],\n",
+    "        cnn_history.history['val_accuracy'][-1]\n",
+    "    ]\n",
+    "})\n",
+    "\n",
+    "print(comparison)\n",
+    "\n",
+    "plt.figure(figsize=(12,5))\n",
+    "plt.subplot(1,2,1)\n",
+    "plt.plot(dnn_history.history['accuracy'], label='DNN Train')\n",
+    "plt.plot(dnn_history.history['val_accuracy'], label='DNN Val')\n",
+    "plt.plot(cnn_history.history['accuracy'], label='CNN Train')\n",
+    "plt.plot(cnn_history.history['val_accuracy'], label='CNN Val')\n",
+    "plt.title('Accuracy')\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Accuracy')\n",
+    "plt.legend()\n",
+    "\n",
+    "plt.subplot(1,2,2)\n",
+    "plt.plot(dnn_history.history['loss'], label='DNN Train')\n",
+    "plt.plot(dnn_history.history['val_loss'], label='DNN Val')\n",
+    "plt.plot(cnn_history.history['loss'], label='CNN Train')\n",
+    "plt.plot(cnn_history.history['val_loss'], label='CNN Val')\n",
+    "plt.title('Loss')\n",
+    "plt.xlabel('Epoch')\n",
+    "plt.ylabel('Loss')\n",
+    "plt.legend()\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "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": "3.12.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

noshot/data/DLE FSD BDA/DLE/DLE 2 (tim stan s)/3 Bounding Boxes.ipynb

@@ -0,0 +1,109 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "1c5ce2ae-0abb-45b4-a94a-7548d9af6b6a",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch\n",
+    "import torchvision\n",
+    "from torchvision.models.detection.faster_rcnn import FastRCNNPredictor\n",
+    "from torchvision.datasets import VOCDetection\n",
+    "from torch.utils.data import DataLoader\n",
+    "import torchvision.transforms as T\n",
+    "import matplotlib.pyplot as plt\n",
+    "import matplotlib.patches as patches\n",
+    "\n",
+    "classes = [\"__background__\", \"apple\", \"banana\", \"orange\"]\n",
+    "\n",
+    "def get_model(num_classes):\n",
+    "    model = torchvision.models.detection.fasterrcnn_resnet50_fpn(weights=\"DEFAULT\")\n",
+    "    in_features = model.roi_heads.box_predictor.cls_score.in_features\n",
+    "    model.roi_heads.box_predictor = FastRCNNPredictor(in_features, num_classes)\n",
+    "    return model\n",
+    "\n",
+    "def collate_fn(batch):\n",
+    "    return tuple(zip(*batch))\n",
+    "\n",
+    "transform = T.Compose([T.ToTensor()])\n",
+    "\n",
+    "def target_transform(target):\n",
+    "    objs = target[\"annotation\"][\"object\"]\n",
+    "    if not isinstance(objs, list):\n",
+    "        objs = [objs]\n",
+    "    boxes = []\n",
+    "    labels = []\n",
+    "    for obj in objs:\n",
+    "        name = obj[\"name\"]\n",
+    "        bbox = obj[\"bndbox\"]\n",
+    "        xmin, ymin, xmax, ymax = int(bbox[\"xmin\"]), int(bbox[\"ymin\"]), int(bbox[\"xmax\"]), int(bbox[\"ymax\"])\n",
+    "        boxes.append([xmin, ymin, xmax, ymax])\n",
+    "        labels.append(classes.index(name) if name in classes else 0)\n",
+    "    return {\"boxes\": torch.tensor(boxes, dtype=torch.float32), \"labels\": torch.tensor(labels, dtype=torch.int64)}\n",
+    "\n",
+    "train_dataset = VOCDetection(\"./\", year=\"2012\", image_set=\"train\", download=True, transform=transform, target_transform=target_transform)\n",
+    "train_loader = DataLoader(train_dataset, batch_size=2, shuffle=True, collate_fn=collate_fn)\n",
+    "\n",
+    "model = get_model(len(classes))\n",
+    "device = torch.device(\"cpu\")\n",
+    "model.to(device)\n",
+    "\n",
+    "params = [p for p in model.parameters() if p.requires_grad]\n",
+    "optimizer = torch.optim.SGD(params, lr=0.005, momentum=0.9, weight_decay=0.0005)\n",
+    "\n",
+    "num_epochs = 1\n",
+    "for epoch in range(num_epochs):\n",
+    "    model.train()\n",
+    "    for images, targets in train_loader:\n",
+    "        images = [img.to(device) for img in images]\n",
+    "        targets = [{k: v.to(device) for k, v in t.items()} for t in targets]\n",
+    "        loss_dict = model(images, targets)\n",
+    "        losses = sum(loss for loss in loss_dict.values())\n",
+    "        optimizer.zero_grad()\n",
+    "        losses.backward()\n",
+    "        optimizer.step()\n",
+    "    print(f\"Epoch {epoch+1}, Loss: {losses.item():.4f}\")\n",
+    "\n",
+    "model.eval()\n",
+    "images, _ = next(iter(train_loader))\n",
+    "img = images[0].to(device)\n",
+    "with torch.no_grad():\n",
+    "    prediction = model([img])\n",
+    "\n",
+    "img_np = img.permute(1, 2, 0).numpy()\n",
+    "fig, ax = plt.subplots(1)\n",
+    "ax.imshow(img_np)\n",
+    "for box, label, score in zip(prediction[0][\"boxes\"], prediction[0][\"labels\"], prediction[0][\"scores\"]):\n",
+    "    if score > 0.5:\n",
+    "        xmin, ymin, xmax, ymax = box\n",
+    "        rect = patches.Rectangle((xmin, ymin), xmax - xmin, ymax - ymin, linewidth=2, edgecolor='r', facecolor='none')\n",
+    "        ax.add_patch(rect)\n",
+    "        ax.text(xmin, ymin, classes[label], bbox=dict(facecolor='yellow', alpha=0.5))\n",
+    "plt.show()\n"
+   ]
+  }
+ ],
+ "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": "3.12.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

noshot/data/DLE FSD BDA/DLE/DLE 2 (tim stan s)/4. FCNN (Image Segmentation).ipynb

@@ -0,0 +1,108 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "b417a942-8300-4101-9851-d65880f3bbb4",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import tensorflow as tf\n",
+    "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n",
+    "from tensorflow.keras.layers import Conv2D, MaxPooling2D, Conv2DTranspose, Input\n",
+    "from tensorflow.keras.models import Model\n",
+    "import matplotlib.pyplot as plt\n",
+    "import numpy as np\n",
+    "\n",
+    "image_dir = \"image path\"\n",
+    "mask_dir = \"mask path\"\n",
+    "\n",
+    "image_datagen = ImageDataGenerator(rescale=1./255)\n",
+    "mask_datagen = ImageDataGenerator(rescale=1./255)\n",
+    "\n",
+    "image_generator = image_datagen.flow_from_directory(\n",
+    "    image_dir,\n",
+    "    class_mode=None,\n",
+    "    color_mode='rgb',\n",
+    "    target_size=(128, 128),\n",
+    "    batch_size=32,\n",
+    "    seed=42\n",
+    ")\n",
+    "\n",
+    "mask_generator = mask_datagen.flow_from_directory(\n",
+    "    mask_dir,\n",
+    "    class_mode=None,\n",
+    "    color_mode='grayscale',\n",
+    "    target_size=(128, 128),\n",
+    "    batch_size=32,\n",
+    "    seed=42\n",
+    ")\n",
+    "\n",
+    "def build_fcnn():\n",
+    "    inputs = Input((128, 128, 3))\n",
+    "    conv1 = Conv2D(128, (3, 3), activation='relu', padding='same')(inputs)\n",
+    "    pool1 = MaxPooling2D((2, 2))(conv1)\n",
+    "\n",
+    "    conv2 = Conv2D(256, (3, 3), activation='relu', padding='same')(pool1)\n",
+    "    pool2 = MaxPooling2D((2, 2))(conv2)\n",
+    "\n",
+    "    conv3 = Conv2D(256, (3, 3), activation='relu', padding='same')(pool2)\n",
+    "    up1 = Conv2DTranspose(128, (2, 2), strides=(2, 2), padding='same')(conv3)\n",
+    "\n",
+    "    conv4 = Conv2D(128, (3, 3), activation='relu', padding='same')(up1)\n",
+    "    up2 = Conv2DTranspose(64, (2, 2), strides=(2, 2), padding='same')(conv4)\n",
+    "\n",
+    "    outputs = Conv2D(1, (1, 1), activation='sigmoid', padding='same')(up2)\n",
+    "    return Model(inputs, outputs)\n",
+    "\n",
+    "model = build_fcnn()\n",
+    "model.compile(optimizer='adam', loss='binary_crossentropy', metrics=['accuracy'])\n",
+    "model.summary()\n",
+    "\n",
+    "def combined_generator(image_gen, mask_gen):\n",
+    "    while True:\n",
+    "        img_batch = next(image_gen)\n",
+    "        mask_batch = next(mask_gen)\n",
+    "        yield img_batch, mask_batch\n",
+    "\n",
+    "train_generator = combined_generator(image_generator, mask_generator)\n",
+    "\n",
+    "model.fit(train_generator, steps_per_epoch=len(image_generator), epochs=50)\n",
+    "\n",
+    "sample_image = image_generator[0][0][0]\n",
+    "predicted_mask = model.predict(np.expand_dims(sample_image, axis=0))[0]\n",
+    "\n",
+    "plt.figure(figsize=(10, 5))\n",
+    "plt.subplot(1, 2, 1)\n",
+    "plt.title(\"Original Image\")\n",
+    "plt.imshow(sample_image)\n",
+    "\n",
+    "plt.subplot(1, 2, 2)\n",
+    "plt.title(\"Predicted Mask\")\n",
+    "plt.imshow(predicted_mask.squeeze(), cmap='gray')\n",
+    "plt.show()"
+   ]
+  }
+ ],
+ "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": "3.12.4"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}