1d_cnn.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "upper-wisdom",
   "metadata": {},
   "outputs": [],
   "source": [
    "from numpy import mean\n",
    "from numpy import std\n",
    "from numpy import dstack\n",
    "from pandas import read_csv\n",
    "from matplotlib import pyplot\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense\n",
    "from keras.layers import Flatten\n",
    "from keras.layers import Dropout\n",
    "from keras.layers.convolutional import Conv1D\n",
    "from keras.layers.convolutional import MaxPooling1D\n",
    "from keras.utils import to_categorical\n",
    "import csv\n",
    "import os\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "concerned-shuttle",
   "metadata": {},
   "outputs": [],
   "source": [
    "def res(arr,x,y):\n",
    "    zeroarray=np.zeros((1,y))\n",
    "    #print(arr.shape)\n",
    "    k=120-x+1\n",
    "    if x<=120:\n",
    "        for i in range(1,k):\n",
    "            arr = np.vstack((arr,zeroarray))      #重設矩陣大小\n",
    "    return arr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "interstate-grain",
   "metadata": {},
   "outputs": [],
   "source": [
    "dirdata1='正手訓/'\n",
    "dirdata2='反手訓/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "apart-purpose",
   "metadata": {},
   "outputs": [],
   "source": [
    "def data_all(dirdata1,dirdata2):\n",
    "    data = []\n",
    "    x=[]\n",
    "    label=[]\n",
    "    max=120\n",
    "    for i in os.listdir(dirdata1):\n",
    "        npy_file = (dirdata1+i)\n",
    "        x=np.load(npy_file)\n",
    "        #x=np.delete(x, [30,31,32,33,34,35,36,37,40,41,42,43,46,47,48,49], axis=1)\n",
    "        #移除15,16,17,18,20,21,23,24點位\n",
    "        a,b=x.shape\n",
    "        x=res(x,a,b)\n",
    "        x = np.array(x)\n",
    "        data.append(x)\n",
    "        label.append(0)\n",
    "    for i in os.listdir(dirdata2):\n",
    "        npy_file = (dirdata2+i)\n",
    "        x=np.load(npy_file)\n",
    "        #x=np.delete(x, [30,31,32,33,34,35,36,37,40,41,42,43,46,47,48,49], axis=1)\n",
    "        a,b=x.shape\n",
    "        x=res(x,a,b)\n",
    "        x = np.array(x)\n",
    "\n",
    "        data.append(x)\n",
    "        label.append(1)\n",
    "    \n",
    "    \n",
    "    data = np.array(data) # 3\n",
    "    label = np.array(label) #3\n",
    "    label = np.eye(2)[label]\n",
    "    \n",
    "    return data,label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "indie-crystal",
   "metadata": {},
   "outputs": [],
   "source": [
    "def data_new(dirdata1,dirdata2):\n",
    "    data = []\n",
    "    x=[]\n",
    "    label=[]\n",
    "    max=120\n",
    "    for i in os.listdir(dirdata1):\n",
    "        npy_file = (dirdata1+i)\n",
    "        x=np.load(npy_file)\n",
    "        x=np.delete(x, [30,31,32,33,34,35,36,37,40,41,42,43,46,47,48,49], axis=1)\n",
    "        #移除15,16,17,18,20,21,23,24點位\n",
    "        a,b=x.shape\n",
    "        x=res(x,a,b)\n",
    "        x = np.array(x)\n",
    "        data.append(x)\n",
    "        label.append(0)\n",
    "    for i in os.listdir(dirdata2):\n",
    "        npy_file = (dirdata2+i)\n",
    "        x=np.load(npy_file)\n",
    "        x=np.delete(x, [30,31,32,33,34,35,36,37,40,41,42,43,46,47,48,49], axis=1)\n",
    "        a,b=x.shape\n",
    "        x=res(x,a,b)\n",
    "        x = np.array(x)\n",
    "\n",
    "        data.append(x)\n",
    "        label.append(1)\n",
    "    \n",
    "    \n",
    "    data = np.array(data) # 3\n",
    "    label = np.array(label) #3\n",
    "    label = np.eye(2)[label]\n",
    "    \n",
    "    return data,label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "antique-charity",
   "metadata": {},
   "outputs": [],
   "source": [
    "data,label=data_all(dirdata1,dirdata2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "approximate-support",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, Y_train, Y_test = train_test_split(data, label, test_size=0.33, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "conservative-dubai",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(440, 120, 50)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "spread-night",
   "metadata": {},
   "outputs": [],
   "source": [
    "count,x,y=data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "reliable-appeal",
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Conv1D(filters=64, kernel_size=3, activation='relu', input_shape=(x,y)))\n",
    "model.add(Conv1D(filters=64, kernel_size=3, activation='relu'))\n",
    "model.add(Dropout(0.5))\n",
    "model.add(MaxPooling1D(pool_size=2))\n",
    "model.add(Flatten())\n",
    "model.add(Dense(100, activation='relu'))\n",
    "model.add(Dense(2, activation='softmax'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "foreign-tutorial",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      "Layer (type)                 Output Shape              Param #   \n",
      "=================================================================\n",
      "conv1d (Conv1D)              (None, 118, 64)           9664      \n",
      "_________________________________________________________________\n",
      "conv1d_1 (Conv1D)            (None, 116, 64)           12352     \n",
      "_________________________________________________________________\n",
      "dropout (Dropout)            (None, 116, 64)           0         \n",
      "_________________________________________________________________\n",
      "max_pooling1d (MaxPooling1D) (None, 58, 64)            0         \n",
      "_________________________________________________________________\n",
      "flatten (Flatten)            (None, 3712)              0         \n",
      "_________________________________________________________________\n",
      "dense (Dense)                (None, 100)               371300    \n",
      "_________________________________________________________________\n",
      "dense_1 (Dense)              (None, 2)                 202       \n",
      "=================================================================\n",
      "Total params: 393,518\n",
      "Trainable params: 393,518\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "invisible-collector",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "8/8 [==============================] - 0s 32ms/step - loss: 699.4298 - accuracy: 0.5617 - val_loss: 9.1935 - val_accuracy: 0.8644\n",
      "Epoch 2/10\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 17.3515 - accuracy: 0.9064 - val_loss: 18.2639 - val_accuracy: 0.9322\n",
      "Epoch 3/10\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 4.1020 - accuracy: 0.9745 - val_loss: 12.4466 - val_accuracy: 0.9322\n",
      "Epoch 4/10\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 2.6903 - accuracy: 0.9957 - val_loss: 0.6509 - val_accuracy: 0.9492\n",
      "Epoch 5/10\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.2998 - accuracy: 0.9915 - val_loss: 1.3842 - val_accuracy: 0.9492\n",
      "Epoch 6/10\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 2.8660e-07 - accuracy: 1.0000 - val_loss: 0.9450 - val_accuracy: 0.9492\n",
      "Epoch 7/10\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.0668 - accuracy: 0.9957 - val_loss: 0.0423 - val_accuracy: 0.9831\n",
      "Epoch 8/10\n",
      "8/8 [==============================] - 0s 8ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
      "Epoch 9/10\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n",
      "Epoch 10/10\n",
      "8/8 [==============================] - 0s 9ms/step - loss: 0.0000e+00 - accuracy: 1.0000 - val_loss: 0.0000e+00 - val_accuracy: 1.0000\n"
     ]
    }
   ],
   "source": [
    "model.compile(loss='categorical_crossentropy', optimizer='adam', metrics=['accuracy'])\n",
    "\n",
    "history = model.fit(X_train,Y_train, epochs=10, validation_split=0.2, batch_size=30, verbose=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "handmade-election",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f, (ax1, ax2) = plt.subplots(1, 2, figsize=(12, 4))\n",
    "t = f.suptitle('class model evaluate', fontsize=12)\n",
    "f.subplots_adjust(top=0.85, wspace=0.3)\n",
    "\n",
    "class_epoch_list = list(range(1,11))\n",
    "ax1.plot(class_epoch_list, history.history['accuracy'], label='class_Train Accuracy')\n",
    "ax1.plot(class_epoch_list, history.history['val_accuracy'], label='class_Validation Accuracy')\n",
    "ax1.set_xticks(np.arange(0, 11, 5))\n",
    "ax1.set_ylabel('class_Accuracy Value')\n",
    "ax1.set_xlabel('class_Epoch')\n",
    "ax1.set_title('class_Accuracy')\n",
    "l1 = ax1.legend(loc=\"best\")\n",
    "\n",
    "ax2.plot(class_epoch_list, history.history['loss'], label='class_Train Loss')\n",
    "ax2.plot(class_epoch_list, history.history['val_loss'], label='class_Validation Loss')\n",
    "ax2.set_xticks(np.arange(0, 11, 5))\n",
    "ax2.set_ylabel('class_Loss Value')\n",
    "ax2.set_xlabel('class_Epoch')\n",
    "ax2.set_title('class_Loss')\n",
    "l2 = ax2.legend(loc=\"best\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "constant-scanner",
   "metadata": {},
   "outputs": [],
   "source": [
    "#model.save('1d_cnn_all.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "signal-opinion",
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.metrics import classification_report"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "desirable-mexican",
   "metadata": {},
   "outputs": [],
   "source": [
    "import keras\n",
    "modello = keras.models.load_model('1d_cnn_all.h5')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "liable-peninsula",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_dirdata1='正手測/'\n",
    "test_dirdata2='反手測/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "civic-lewis",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_data,test_label=data_all(test_dirdata1,test_dirdata2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "considerable-error",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 - 0s\n"
     ]
    }
   ],
   "source": [
    "time_pre=modello.predict(test_data,verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "revised-mining",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1., 0.],\n",
       "       [1., 0.],\n",
       "       [1., 0.]])"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_label[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "systematic-lunch",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 1.],\n",
       "       [0., 1.],\n",
       "       [0., 1.]], dtype=float32)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "time_pre[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "introductory-mumbai",
   "metadata": {},
   "outputs": [],
   "source": [
    "Y_test = np.argmax(test_label,axis=1)\n",
    "Y_pred = np.argmax(time_pre,axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "outside-survival",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "mat = confusion_matrix(Y_test,Y_pred)\n",
    "\n",
    "sns.set(font_scale=1.5)\n",
    "plt.subplot(1,2,1)\n",
    "sns.heatmap(mat, square=True, annot=True, fmt='d', cbar=False)\n",
    "plt.xticks(fontsize=20)\n",
    "plt.yticks(fontsize=20)\n",
    "\n",
    "plt.ylabel('true label')\n",
    "plt.xlabel('predicted label');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "drawn-sight",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "    forehand       1.00      0.94      0.97       159\n",
      "    backhand       0.94      1.00      0.97       145\n",
      "\n",
      "    accuracy                           0.97       304\n",
      "   macro avg       0.97      0.97      0.97       304\n",
      "weighted avg       0.97      0.97      0.97       304\n",
      "\n"
     ]
    }
   ],
   "source": [
    "target_names = [\"forehand\",\"backhand\"]\n",
    "print(classification_report(Y_test,Y_pred,target_names=target_names))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "danish-coupon",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Predicted label (first 100)')"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 288x216 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(4,3))\n",
    "\n",
    "\n",
    "# plt.subplot(3,1,1)\n",
    "plt.plot(time_pre[:500], color = 'red', label = 'Predicted')\n",
    "plt.plot(Y_test[:500], color = 'black', linewidth=2, label = 'True value')\n",
    "\n",
    "plt.legend(frameon=False)\n",
    "plt.ylabel(\"label forecast\")\n",
    "plt.xlabel(\"time\")\n",
    "plt.title(\"Predicted label (first 100)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "gorgeous-repair",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}