SW-Wiktor-Bombola/SW-Unity/Plants Neural Network.ipynb
2021-12-19 23:21:08 +01:00

387 lines
13 KiB
Plaintext

{
"cells": [
{
"cell_type": "code",
"execution_count": 52,
"id": "comprehensive-talent",
"metadata": {},
"outputs": [],
"source": [
"import cv2\n",
"import os\n",
"import numpy as np\n",
"import tensorflow as tf\n",
"from tensorflow.keras.models import Sequential\n",
"from tensorflow.keras.layers import Dense, Dropout, Flatten, Activation, Conv2D, MaxPooling2D\n",
"from sklearn.neural_network import MLPClassifier\n",
"from sklearn.model_selection import train_test_split\n",
"from sklearn.metrics import classification_report\n",
"import re"
]
},
{
"cell_type": "code",
"execution_count": 26,
"id": "sapphire-monte",
"metadata": {},
"outputs": [],
"source": [
"def preprocessing(image):\n",
" scale_percent = 10\n",
" width = int(image.shape[1] * scale_percent / 100)\n",
" height = int(image.shape[0] * scale_percent / 100)\n",
" dim = (width, height)\n",
" return cv2.resize(image, dim, interpolation = cv2.INTER_AREA)\n",
"\n",
"\n",
"def read_data(data_images):\n",
" x, y = [], []\n",
" for image in data_images:\n",
" img = cv2.imread(image, cv2.IMREAD_COLOR)\n",
" img = preprocessing(img)\n",
" y_label = re.search(r\"(?<=-).(?=-)\", image).group(0)\n",
" x.append(img)\n",
" y.append(y_label)\n",
" return x, y"
]
},
{
"cell_type": "code",
"execution_count": 6,
"id": "greenhouse-needle",
"metadata": {},
"outputs": [],
"source": [
"location = \"capturedframe/\"\n",
"data_images = os.listdir(location)\n",
"# for x in data_images:\n",
"# os.rename(location+x, \"tree-1-\"+ x[13:])\n",
"data_images = [location + x for x in data_images if x.endswith(\".png\")]\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "black-channel",
"metadata": {},
"outputs": [],
"source": [
"print()"
]
},
{
"cell_type": "code",
"execution_count": 28,
"id": "built-palestinian",
"metadata": {},
"outputs": [],
"source": [
"x, y = read_data(data_images)"
]
},
{
"cell_type": "code",
"execution_count": 73,
"id": "amber-wisconsin",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"['1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1', '1']\n"
]
}
],
"source": [
"print(y)"
]
},
{
"cell_type": "code",
"execution_count": 31,
"id": "instant-frequency",
"metadata": {},
"outputs": [],
"source": [
"X_train, X_test, y_train, y_test = train_test_split(x,y, test_size=0.2, random_state=81)"
]
},
{
"cell_type": "code",
"execution_count": 37,
"id": "dried-college",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(60, 80, 3)"
]
},
"execution_count": 37,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X_train[0].shape"
]
},
{
"cell_type": "code",
"execution_count": 38,
"id": "tutorial-interpretation",
"metadata": {},
"outputs": [],
"source": [
"X_train = np.array([x / 255.0 for x in X_train], dtype=np.float64)\n",
"X_test = np.array([x / 255.0 for x in X_test], dtype=np.float64)\n",
"\n",
"y_train = np.array(y_train, dtype=np.int64)\n",
"y_test = np.array(y_test, dtype=np.int64)\n"
]
},
{
"cell_type": "code",
"execution_count": 43,
"id": "green-being",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"[[[0.00073818 0.00086121 0.00070742]\n",
" [0.0009381 0.00112265 0.0009381 ]\n",
" [0.00104575 0.00129181 0.00107651]\n",
" ...\n",
" [0.00246059 0.00273741 0.00247597]\n",
" [0.00229143 0.00267589 0.00241446]\n",
" [0.00232218 0.00276817 0.00247597]]\n",
"\n",
" [[0.00089196 0.00099962 0.00081507]\n",
" [0.00107651 0.00130719 0.00109189]\n",
" [0.0009381 0.00112265 0.0009381 ]\n",
" ...\n",
" [0.00244521 0.00276817 0.00250673]\n",
" [0.00218378 0.00270665 0.0023837 ]\n",
" [0.00219915 0.002599 0.0023837 ]]\n",
"\n",
" [[0.0012303 0.00124567 0.00103037]\n",
" [0.00113802 0.00132257 0.00110727]\n",
" [0.00099962 0.0012303 0.00103037]\n",
" ...\n",
" [0.00233756 0.00279892 0.00249135]\n",
" [0.00226067 0.00264514 0.00232218]\n",
" [0.00226067 0.00267589 0.00236832]]\n",
"\n",
" ...\n",
"\n",
" [[0.00084583 0.00101499 0.00083045]\n",
" [0.00090734 0.00112265 0.00092272]\n",
" [0.00090734 0.00109189 0.00089196]\n",
" ...\n",
" [0.00229143 0.00292195 0.002599 ]\n",
" [0.00210688 0.00255286 0.00224529]\n",
" [0.00226067 0.00270665 0.00250673]]\n",
"\n",
" [[0.00087659 0.00101499 0.00079969]\n",
" [0.00079969 0.0009381 0.00075356]\n",
" [0.00089196 0.00107651 0.00089196]\n",
" ...\n",
" [0.00247597 0.00290657 0.00264514]\n",
" [0.00236832 0.00270665 0.00246059]\n",
" [0.00235294 0.00293733 0.002599 ]]\n",
"\n",
" [[0.0009381 0.00112265 0.00092272]\n",
" [0.00084583 0.00099962 0.00079969]\n",
" [0.00084583 0.00099962 0.00081507]\n",
" ...\n",
" [0.00282968 0.00315263 0.00290657]\n",
" [0.00276817 0.0031065 0.0028912 ]\n",
" [0.00224529 0.00278354 0.00230681]]]\n"
]
}
],
"source": [
"print((X_train[0]))"
]
},
{
"cell_type": "code",
"execution_count": 84,
"id": "natural-cutting",
"metadata": {},
"outputs": [],
"source": [
"model = Sequential()"
]
},
{
"cell_type": "code",
"execution_count": 85,
"id": "conservative-hypothetical",
"metadata": {},
"outputs": [],
"source": [
"model.add(Conv2D(32, (3,3), activation='relu', input_shape=(X_train[0].shape)))\n",
"model.add(MaxPooling2D((2,2)))\n",
"model.add(Conv2D(64, (3,3), activation='relu'))\n",
"model.add(MaxPooling2D((2,2)))\n",
"\n",
"model.add(Conv2D(32, (3,3), activation='relu'))\n",
"\n",
"model.add(MaxPooling2D((2,2)))\n",
"model.add(Flatten())\n",
"model.add(Dense(256, activation='relu'))\n",
"model.add(Dense(2, activation='sigmoid'))"
]
},
{
"cell_type": "code",
"execution_count": 86,
"id": "illegal-zoning",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Model: \"sequential_6\"\n",
"_________________________________________________________________\n",
"Layer (type) Output Shape Param # \n",
"=================================================================\n",
"conv2d_16 (Conv2D) (None, 58, 78, 32) 896 \n",
"_________________________________________________________________\n",
"max_pooling2d_12 (MaxPooling (None, 29, 39, 32) 0 \n",
"_________________________________________________________________\n",
"conv2d_17 (Conv2D) (None, 27, 37, 64) 18496 \n",
"_________________________________________________________________\n",
"max_pooling2d_13 (MaxPooling (None, 13, 18, 64) 0 \n",
"_________________________________________________________________\n",
"conv2d_18 (Conv2D) (None, 11, 16, 32) 18464 \n",
"_________________________________________________________________\n",
"max_pooling2d_14 (MaxPooling (None, 5, 8, 32) 0 \n",
"_________________________________________________________________\n",
"flatten_6 (Flatten) (None, 1280) 0 \n",
"_________________________________________________________________\n",
"dense_12 (Dense) (None, 256) 327936 \n",
"_________________________________________________________________\n",
"dense_13 (Dense) (None, 2) 514 \n",
"=================================================================\n",
"Total params: 366,306\n",
"Trainable params: 366,306\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n",
"None\n"
]
}
],
"source": [
"print(model.summary())"
]
},
{
"cell_type": "code",
"execution_count": 87,
"id": "cardiac-highland",
"metadata": {},
"outputs": [],
"source": [
"model.compile(optimizer='adam',\n",
" loss=tf.keras.losses.SparseCategoricalCrossentropy(),\n",
" metrics=['accuracy'])"
]
},
{
"cell_type": "code",
"execution_count": 88,
"id": "informed-baker",
"metadata": {
"scrolled": false
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/10\n",
"9/9 [==============================] - 1s 62ms/step - loss: 0.4567 - accuracy: 0.9173 - val_loss: 0.0150 - val_accuracy: 1.0000\n",
"Epoch 2/10\n",
"9/9 [==============================] - 0s 52ms/step - loss: 0.0021 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 3/10\n",
"9/9 [==============================] - 0s 50ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 4/10\n",
"9/9 [==============================] - 0s 50ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 5/10\n",
"9/9 [==============================] - 0s 51ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 6/10\n",
"9/9 [==============================] - 0s 50ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 7/10\n",
"9/9 [==============================] - 0s 53ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 8/10\n",
"9/9 [==============================] - 0s 52ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 9/10\n",
"9/9 [==============================] - 0s 50ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
"Epoch 10/10\n",
"9/9 [==============================] - 0s 49ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n"
]
}
],
"source": [
"history = model.fit(X_train, y_train, epochs=10,\n",
" validation_data=(X_test, y_test))"
]
},
{
"cell_type": "code",
"execution_count": 72,
"id": "inclusive-chess",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"3/3 - 0s - loss: 0.0000e+00 - accuracy: 1.0000\n"
]
}
],
"source": [
"test_loss, test_acc = model.evaluate(X_test, y_test, verbose=2)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "marine-satellite",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"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": "3.9.1"
}
},
"nbformat": 4,
"nbformat_minor": 5
}