s473575/inteligentna_smieciarka
1
1
Fork 1
Code Issues Pull Requests Projects Releases Wiki Activity
b78f57386e
inteligentna_smieciarka/trash-classification/trash-prediction.ipynb

710 lines
1.7 MiB
Plaintext
Raw Normal View History

80% accuracy
2023-06-02 21:55:44 +02:00
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "6db5fd63",
"metadata": {},
"outputs": [],
"source": [
"import matplotlib.pyplot as plt\n",
"import numpy as np\n",
"import os\n",
"import PIL\n",
"import tensorflow as tf\n",
"\n",
"from tensorflow import keras\n",
"from tensorflow.keras import layers\n",
"from tensorflow.keras.models import Sequential"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "2cb154b9",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[]"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tf.config.list_physical_devices('GPU')"
]
},
{
"cell_type": "code",
"execution_count": 3,
"id": "4edd961f",
"metadata": {},
"outputs": [
{
"ename": "KeyboardInterrupt",
"evalue": "",
"output_type": "error",
"traceback": [
"\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
"\u001b[1;31mKeyboardInterrupt\u001b[0m Traceback (most recent call last)",
"Cell \u001b[1;32mIn[3], line 10\u001b[0m\n\u001b[0;32m 8\u001b[0m image_path \u001b[38;5;241m=\u001b[39m os\u001b[38;5;241m.\u001b[39mpath\u001b[38;5;241m.\u001b[39mjoin(data_dir, image_class, image)\n\u001b[0;32m 9\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m: \n\u001b[1;32m---> 10\u001b[0m img \u001b[38;5;241m=\u001b[39m \u001b[43mcv2\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mimread\u001b[49m\u001b[43m(\u001b[49m\u001b[43mimage_path\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m 11\u001b[0m tip \u001b[38;5;241m=\u001b[39m imghdr\u001b[38;5;241m.\u001b[39mwhat(image_path)\n\u001b[0;32m 12\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m tip \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m image_exts: \n",
"\u001b[1;31mKeyboardInterrupt\u001b[0m: "
]
}
],
"source": [
"import cv2\n",
"import imghdr\n",
"import os\n",
"data_dir = 'trash/validate' \n",
"image_exts = ['jpeg','jpg', 'bmp', 'png']\n",
"for image_class in os.listdir(data_dir): \n",
" for image in os.listdir(os.path.join(data_dir, image_class)):\n",
" image_path = os.path.join(data_dir, image_class, image)\n",
" try: \n",
" img = cv2.imread(image_path)\n",
" tip = imghdr.what(image_path)\n",
" if tip not in image_exts: \n",
" print('Image not in ext list {}'.format(image_path))\n",
" os.remove(image_path)\n",
" except Exception as e: \n",
" print('Issue with image {}'.format(image_path))\n",
" # os.remove(image_path)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"id": "a82eb04c",
"metadata": {},
"outputs": [],
"source": [
"batch_size = 32\n",
"img_height = 180\n",
"img_width = 180"
]
},
{
"cell_type": "code",
"execution_count": 5,
"id": "1bde4009",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Found 7524 files belonging to 4 classes.\n",
"Found 1531 files belonging to 4 classes.\n",
"Found 2562 files belonging to 4 classes.\n"
]
}
],
"source": [
"train_ds = tf.keras.utils.image_dataset_from_directory(\n",
" \"trash/train\",\n",
" image_size = (img_height, img_width),\n",
" seed = 123,\n",
" batch_size = batch_size\n",
")\n",
"val_ds = tf.keras.utils.image_dataset_from_directory(\n",
" \"trash/validate\",\n",
" image_size = (img_height, img_width),\n",
" seed = 123,\n",
" batch_size = batch_size\n",
")\n",
"test_ds = tf.keras.utils.image_dataset_from_directory(\n",
" \"trash/test\",\n",
" image_size = (img_height, img_width),\n",
" seed = 123,\n",
" batch_size = batch_size\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "f4fe3a65",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['biological', 'glass', 'paper', 'plastic']\n"
]
}
],
"source": [
"class_names = train_ds.class_names\n",
"print(class_names)"
]
},
{
"cell_type": "code",
"execution_count": 7,
"id": "3bbea143",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 1000x1000 with 9 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"import matplotlib.pyplot as plt\n",
"\n",
"plt.figure(figsize=(10, 10))\n",
"for images, labels in train_ds.take(1):\n",
" for i in range(9):\n",
" ax = plt.subplot(3, 3, i + 1)\n",
" plt.imshow(images[i].numpy().astype(\"uint8\"))\n",
" plt.title(class_names[labels[i]])\n",
" plt.axis(\"off\")"
]
},
{
"cell_type": "code",
"execution_count": 8,
"id": "ae666555",
"metadata": {},
"outputs": [],
"source": [
"AUTOTUNE = tf.data.AUTOTUNE\n",
"\n",
"train_ds = train_ds.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)\n",
"val_ds = val_ds.cache().prefetch(buffer_size=AUTOTUNE)"
]
},
{
"cell_type": "code",
"execution_count": 9,
"id": "aadbc1b5",
"metadata": {},
"outputs": [],
"source": [
"normalization_layer = layers.Rescaling(1./255)"
]
},
{
"cell_type": "code",
"execution_count": 10,
"id": "e515e074",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"0.0 1.0\n"
]
}
],
"source": [
"normalized_ds = train_ds.map(lambda x, y: (normalization_layer(x), y))\n",
"image_batch, labels_batch = next(iter(normalized_ds))\n",
"first_image = image_batch[0]\n",
"# Notice the pixel values are now in `[0,1]`.\n",
"print(np.min(first_image), np.max(first_image))"
]
},
{
"cell_type": "code",
"execution_count": 11,
"id": "6d8901bf",
"metadata": {},
"outputs": [],
"source": [
"num_classes = len(class_names)\n",
"\n",
"model = Sequential([\n",
" layers.Rescaling(1./255, input_shape=(img_height, img_width, 3)),\n",
" layers.Conv2D(16, 3, padding='same', activation='relu'),\n",
" layers.MaxPooling2D(),\n",
" layers.Conv2D(32, 3, padding='same', activation='relu'),\n",
" layers.MaxPooling2D(),\n",
" layers.Conv2D(64, 3, padding='same', activation='relu'),\n",
" layers.MaxPooling2D(),\n",
" layers.Flatten(),\n",
" layers.Dense(128, activation='relu'),\n",
" layers.Dense(num_classes)\n",
"])"
]
},
{
"cell_type": "code",
"execution_count": 12,
"id": "13ace97f",
"metadata": {},
"outputs": [],
"source": [
"model.compile(optimizer='adam',\n",
" loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "code",
"execution_count": 13,
"id": "8a493ba3",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model: \"sequential\"\n",
"_________________________________________________________________\n",
" Layer (type) Output Shape Param # \n",
"=================================================================\n",
" rescaling_1 (Rescaling) (None, 180, 180, 3) 0 \n",
" \n",
" conv2d (Conv2D) (None, 180, 180, 16) 448 \n",
" \n",
" max_pooling2d (MaxPooling2D (None, 90, 90, 16) 0 \n",
" ) \n",
" \n",
" conv2d_1 (Conv2D) (None, 90, 90, 32) 4640 \n",
" \n",
" max_pooling2d_1 (MaxPooling (None, 45, 45, 32) 0 \n",
" 2D) \n",
" \n",
" conv2d_2 (Conv2D) (None, 45, 45, 64) 18496 \n",
" \n",
" max_pooling2d_2 (MaxPooling (None, 22, 22, 64) 0 \n",
" 2D) \n",
" \n",
" flatten (Flatten) (None, 30976) 0 \n",
" \n",
" dense (Dense) (None, 128) 3965056 \n",
" \n",
" dense_1 (Dense) (None, 4) 516 \n",
" \n",
"=================================================================\n",
"Total params: 3,989,156\n",
"Trainable params: 3,989,156\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
]
}
],
"source": [
"model.summary()"
]
},
{
"cell_type": "code",
"execution_count": 14,
"id": "ab563098",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"236/236 [==============================] - 101s 426ms/step - loss: 0.9672 - accuracy: 0.5795 - val_loss: 0.8036 - val_accuracy: 0.6656\n",
"Epoch 2/10\n",
"236/236 [==============================] - 118s 499ms/step - loss: 0.6830 - accuracy: 0.7228 - val_loss: 0.6583 - val_accuracy: 0.7466\n",
"Epoch 3/10\n",
"236/236 [==============================] - 103s 433ms/step - loss: 0.5218 - accuracy: 0.7920 - val_loss: 0.5555 - val_accuracy: 0.7884\n",
"Epoch 4/10\n",
"236/236 [==============================] - 98s 415ms/step - loss: 0.3826 - accuracy: 0.8542 - val_loss: 0.6888 - val_accuracy: 0.7688\n",
"Epoch 5/10\n",
"236/236 [==============================] - 97s 413ms/step - loss: 0.2760 - accuracy: 0.8970 - val_loss: 0.5302 - val_accuracy: 0.8295\n",
"Epoch 6/10\n",
"236/236 [==============================] - 97s 410ms/step - loss: 0.1951 - accuracy: 0.9269 - val_loss: 0.6208 - val_accuracy: 0.8184\n",
"Epoch 7/10\n",
"236/236 [==============================] - 96s 408ms/step - loss: 0.1234 - accuracy: 0.9545 - val_loss: 0.6179 - val_accuracy: 0.8354\n",
"Epoch 8/10\n",
"236/236 [==============================] - 98s 414ms/step - loss: 0.0947 - accuracy: 0.9680 - val_loss: 0.6647 - val_accuracy: 0.8406\n",
"Epoch 9/10\n",
"236/236 [==============================] - 103s 436ms/step - loss: 0.0624 - accuracy: 0.9805 - val_loss: 0.7285 - val_accuracy: 0.8321\n",
"Epoch 10/10\n",
"236/236 [==============================] - 99s 421ms/step - loss: 0.0649 - accuracy: 0.9793 - val_loss: 0.8687 - val_accuracy: 0.8374\n"
]
}
],
"source": [
"epochs=10\n",
"history = model.fit(\n",
" train_ds,\n",
" validation_data=val_ds,\n",
" epochs=epochs\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 15,
"id": "2456bf81",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 800x800 with 2 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"acc = history.history['accuracy']\n",
"val_acc = history.history['val_accuracy']\n",
"\n",
"loss = history.history['loss']\n",
"val_loss = history.history['val_loss']\n",
"\n",
"epochs_range = range(epochs)\n",
"\n",
"plt.figure(figsize=(8, 8))\n",
"plt.subplot(1, 2, 1)\n",
"plt.plot(epochs_range, acc, label='Training Accuracy')\n",
"plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
"plt.legend(loc='lower right')\n",
"plt.title('Training and Validation Accuracy')\n",
"\n",
"plt.subplot(1, 2, 2)\n",
"plt.plot(epochs_range, loss, label='Training Loss')\n",
"plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
"plt.legend(loc='upper right')\n",
"plt.title('Training and Validation Loss')\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 16,
"id": "8b5af9af",
"metadata": {},
"outputs": [],
"source": [
"data_augmentation = keras.Sequential(\n",
" [\n",
" layers.RandomFlip(\"horizontal\",\n",
" input_shape=(img_height,\n",
" img_width,\n",
" 3)),\n",
" layers.RandomRotation(0.1),\n",
" layers.RandomZoom(0.1),\n",
" ]\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 17,
"id": "1873e170",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 1000x1000 with 9 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"plt.figure(figsize=(10, 10))\n",
"for images, _ in train_ds.take(1):\n",
" for i in range(9):\n",
" augmented_images = data_augmentation(images)\n",
" ax = plt.subplot(3, 3, i + 1)\n",
" plt.imshow(augmented_images[0].numpy().astype(\"uint8\"))\n",
" plt.axis(\"off\")"
]
},
{
"cell_type": "code",
"execution_count": 18,
"id": "17833614",
"metadata": {},
"outputs": [],
"source": [
"model = Sequential([\n",
" data_augmentation,\n",
" layers.Rescaling(1./255),\n",
" layers.Conv2D(16, 3, padding='same', activation='relu'),\n",
" layers.MaxPooling2D(),\n",
" layers.Conv2D(32, 3, padding='same', activation='relu'),\n",
" layers.MaxPooling2D(),\n",
" layers.Conv2D(64, 3, padding='same', activation='relu'),\n",
" layers.MaxPooling2D(),\n",
" layers.Dropout(0.2),\n",
" layers.Flatten(),\n",
" layers.Dense(128, activation='relu'),\n",
" layers.Dense(num_classes)\n",
"])"
]
},
{
"cell_type": "code",
"execution_count": 19,
"id": "ad1fd3fe",
"metadata": {},
"outputs": [],
"source": [
"model.compile(optimizer='adam',\n",
" loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "code",
"execution_count": 20,
"id": "51028e4e",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model: \"sequential_2\"\n",
"_________________________________________________________________\n",
" Layer (type) Output Shape Param # \n",
"=================================================================\n",
" sequential_1 (Sequential) (None, 180, 180, 3) 0 \n",
" \n",
" rescaling_2 (Rescaling) (None, 180, 180, 3) 0 \n",
" \n",
" conv2d_3 (Conv2D) (None, 180, 180, 16) 448 \n",
" \n",
" max_pooling2d_3 (MaxPooling (None, 90, 90, 16) 0 \n",
" 2D) \n",
" \n",
" conv2d_4 (Conv2D) (None, 90, 90, 32) 4640 \n",
" \n",
" max_pooling2d_4 (MaxPooling (None, 45, 45, 32) 0 \n",
" 2D) \n",
" \n",
" conv2d_5 (Conv2D) (None, 45, 45, 64) 18496 \n",
" \n",
" max_pooling2d_5 (MaxPooling (None, 22, 22, 64) 0 \n",
" 2D) \n",
" \n",
" dropout (Dropout) (None, 22, 22, 64) 0 \n",
" \n",
" flatten_1 (Flatten) (None, 30976) 0 \n",
" \n",
" dense_2 (Dense) (None, 128) 3965056 \n",
" \n",
" dense_3 (Dense) (None, 4) 516 \n",
" \n",
"=================================================================\n",
"Total params: 3,989,156\n",
"Trainable params: 3,989,156\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n"
]
}
],
"source": [
"model.summary()\n"
]
},
{
"cell_type": "code",
"execution_count": 21,
"id": "d494a467",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/15\n",
"236/236 [==============================] - 109s 459ms/step - loss: 1.0434 - accuracy: 0.5354 - val_loss: 0.8396 - val_accuracy: 0.6251\n",
"Epoch 2/15\n",
"236/236 [==============================] - 109s 463ms/step - loss: 0.8393 - accuracy: 0.6411 - val_loss: 0.7897 - val_accuracy: 0.6558\n",
"Epoch 3/15\n",
"236/236 [==============================] - 117s 496ms/step - loss: 0.7672 - accuracy: 0.6809 - val_loss: 0.7359 - val_accuracy: 0.7035\n",
"Epoch 4/15\n",
"236/236 [==============================] - 120s 508ms/step - loss: 0.7099 - accuracy: 0.7027 - val_loss: 0.7405 - val_accuracy: 0.6871\n",
"Epoch 5/15\n",
"236/236 [==============================] - 115s 489ms/step - loss: 0.6702 - accuracy: 0.7240 - val_loss: 0.7335 - val_accuracy: 0.7211\n",
"Epoch 6/15\n",
"236/236 [==============================] - 109s 461ms/step - loss: 0.6064 - accuracy: 0.7582 - val_loss: 0.7130 - val_accuracy: 0.7302\n",
"Epoch 7/15\n",
"236/236 [==============================] - 109s 460ms/step - loss: 0.5666 - accuracy: 0.7727 - val_loss: 0.6493 - val_accuracy: 0.7564\n",
"Epoch 8/15\n",
"236/236 [==============================] - 110s 464ms/step - loss: 0.5299 - accuracy: 0.7939 - val_loss: 0.5655 - val_accuracy: 0.7936\n",
"Epoch 9/15\n",
"236/236 [==============================] - 109s 460ms/step - loss: 0.5043 - accuracy: 0.8069 - val_loss: 0.5580 - val_accuracy: 0.7956\n",
"Epoch 10/15\n",
"236/236 [==============================] - 109s 460ms/step - loss: 0.4594 - accuracy: 0.8266 - val_loss: 0.6662 - val_accuracy: 0.7720\n",
"Epoch 11/15\n",
"236/236 [==============================] - 111s 471ms/step - loss: 0.4437 - accuracy: 0.8359 - val_loss: 0.5410 - val_accuracy: 0.8034\n",
"Epoch 12/15\n",
"236/236 [==============================] - 116s 492ms/step - loss: 0.4287 - accuracy: 0.8381 - val_loss: 0.5495 - val_accuracy: 0.8106\n",
"Epoch 13/15\n",
"236/236 [==============================] - 126s 534ms/step - loss: 0.3930 - accuracy: 0.8511 - val_loss: 0.5732 - val_accuracy: 0.8027\n",
"Epoch 14/15\n",
"236/236 [==============================] - 114s 481ms/step - loss: 0.3790 - accuracy: 0.8624 - val_loss: 0.5392 - val_accuracy: 0.8197\n",
"Epoch 15/15\n",
"236/236 [==============================] - 111s 468ms/step - loss: 0.3558 - accuracy: 0.8650 - val_loss: 0.5117 - val_accuracy: 0.8217\n"
]
}
],
"source": [
"epochs = 15\n",
"history = model.fit(\n",
" train_ds,\n",
" validation_data=val_ds,\n",
" epochs=epochs\n",
")"
]
},
{
"cell_type": "code",
"execution_count": 23,
"id": "5f971072",
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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
"text/plain": [
"<Figure size 800x800 with 2 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"acc = history.history['accuracy']\n",
"val_acc = history.history['val_accuracy']\n",
"\n",
"loss = history.history['loss']\n",
"val_loss = history.history['val_loss']\n",
"\n",
"epochs_range = range(epochs)\n",
"\n",
"plt.figure(figsize=(8, 8))\n",
"plt.subplot(1, 2, 1)\n",
"plt.plot(epochs_range, acc, label='Training Accuracy')\n",
"plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
"plt.legend(loc='lower right')\n",
"plt.title('Training and Validation Accuracy')\n",
"\n",
"plt.subplot(1, 2, 2)\n",
"plt.plot(epochs_range, loss, label='Training Loss')\n",
"plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
"plt.legend(loc='upper right')\n",
"plt.title('Training and Validation Loss')\n",
"plt.show()"
]
},
{
"cell_type": "code",
"execution_count": 22,
"id": "1c65bca1",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"81/81 [==============================] - 9s 110ms/step - loss: 0.6138 - accuracy: 0.8009\n"
]
},
{
"data": {
"text/plain": [
"[0.6138237714767456, 0.800936758518219]"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model.evaluate(test_ds)"
]
},
{
"cell_type": "code",
"execution_count": 25,
"id": "3b3a2363",
"metadata": {},
"outputs": [],
"source": [
"model.save(\"m_with_data_augmentation.h5\")"
]
},
{
"cell_type": "code",
"execution_count": 34,
"id": "9013d852",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"1/1 [==============================] - 0s 22ms/step\n",
"This image most likely belongs to biological with a 90.06 percent confidence.\n"
]
}
],
"source": [
"sunflower_path = \"trash/test/biological/biological476.jpg\"\n",
"\n",
"img = tf.keras.utils.load_img(\n",
" sunflower_path, target_size=(img_height, img_width)\n",
")\n",
"img_array = tf.keras.utils.img_to_array(img)\n",
"img_array = tf.expand_dims(img_array, 0) # Create a batch\n",
"\n",
"predictions = model.predict(img_array)\n",
"score = tf.nn.softmax(predictions[0])\n",
"\n",
"print(\n",
" \"This image most likely belongs to {} with a {:.2f} percent confidence.\"\n",
" .format(class_names[np.argmax(score)], 100 * np.max(score))\n",
")"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "myenv",
"language": "python",
"name": "myenv"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.13"
}
},
"nbformat": 4,
"nbformat_minor": 5
}
Reference in New Issue Copy Permalink