├── Ann.ipynb ├── LSTM.ipynb ├── algos.py ├── aqi.csv ├── aqi.py ├── aqi.pyc ├── aqivar.py ├── dframe.pkl ├── final report.pdf ├── final_aqi.pkl ├── main.py ├── pollution.csv ├── preprocessing.py └── svr.ipynb /Ann.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 1, 6 | "metadata": {}, 7 | "outputs": [], 8 | "source": [ 9 | "import tensorflow as tf\n", 10 | "import numpy as np\n", 11 | "from tensorflow import keras\n", 12 | "import pandas as pd\n", 13 | "from sklearn.preprocessing import LabelEncoder\n", 14 | "from matplotlib import pyplot as plt\n", 15 | "from sklearn.metrics import mean_squared_error\n", 16 | "from pandas import DataFrame\n", 17 | "from pandas import concat\n", 18 | "from sklearn.preprocessing import MinMaxScaler\n", 19 | "from sklearn.metrics import r2_score\n", 20 | "import time\n", 21 | "\n", 22 | "\n", 23 | "\n", 24 | "\n", 25 | "def series_to_supervised(data, n_in=1, n_out=1, dropnan=True):\n", 26 | " n_vars = 1 if type(data) is list else data.shape[1]\n", 27 | " df = DataFrame(data)\n", 28 | " cols, names = list(), list()\n", 29 | " # input sequence (t-n, ... t-1)\n", 30 | " for i in range(n_in, 0, -1):\n", 31 | " cols.append(df.shift(i))\n", 32 | " names += [('var%d(t-%d)' % (j+1, i)) for j in range(n_vars)]\n", 33 | " # forecast sequence (t, t+1)\n", 34 | " for i in range(0, n_out):\n", 35 | " cols.append(df.shift(-i))\n", 36 | " if i == 0:\n", 37 | " names += [('var%d(t)' % (j+1)) for j in range(n_vars)]\n", 38 | " else:\n", 39 | " names += [('var%d(t+%d)' % (j+1, i)) for j in range(n_vars)]\n", 40 | " # put it all together\n", 41 | " agg = concat(cols, axis=1)\n", 42 | " agg.columns = names\n", 43 | " # drop rows with NaN values\n", 44 | " if dropnan:\n", 45 | " agg.dropna(inplace=True)\n", 46 | " return agg\n" 47 | ] 48 | }, 49 | { 50 | "cell_type": "code", 51 | "execution_count": 2, 52 | "metadata": {}, 53 | "outputs": [ 54 | { 55 | "name": "stdout", 56 | "output_type": "stream", 57 | "text": [ 58 | " var1(t-1) var2(t-1) var3(t-1) var4(t-1) var5(t-1) var6(t-1) \\\n", 59 | "1 129.0 -16.0 -4.0 1020.0 2.0 1.79 \n", 60 | "2 148.0 -15.0 -4.0 1020.0 2.0 2.68 \n", 61 | "3 159.0 -11.0 -5.0 1021.0 2.0 3.57 \n", 62 | "4 181.0 -7.0 -5.0 1022.0 2.0 5.36 \n", 63 | "5 138.0 -7.0 -5.0 1022.0 2.0 6.25 \n", 64 | "\n", 65 | " var7(t-1) var8(t-1) var1(t) \n", 66 | "1 0.0 0.0 148.0 \n", 67 | "2 0.0 0.0 159.0 \n", 68 | "3 0.0 0.0 181.0 \n", 69 | "4 1.0 0.0 138.0 \n", 70 | "5 2.0 0.0 109.0 \n" 71 | ] 72 | } 73 | ], 74 | "source": [ 75 | "dataset = pd.read_csv('pollution.csv', header=0, index_col=0)\n", 76 | "values = dataset.values\n", 77 | "# integer encode direction\n", 78 | "encoder = LabelEncoder()\n", 79 | "values[:,4] = encoder.fit_transform(values[:,4])\n", 80 | "# ensure all data is float\n", 81 | "values = values.astype('float32')\n", 82 | "reframed = series_to_supervised(values, 1, 1)\n", 83 | "# drop columns we don't want to predict\n", 84 | "reframed.drop(reframed.columns[[-7,-6,-5,-4,-3,-2,-1]], axis=1, inplace=True)\n", 85 | "print(reframed.head())\n", 86 | "\n", 87 | "values = reframed.values\n", 88 | "scaler = MinMaxScaler(feature_range=(0, 1), copy=True)\n", 89 | "scaled_features = scaler.fit_transform(values[:,:-1])\n", 90 | "scaled_label = scaler.fit_transform(values[:,-1].reshape(-1,1))\n", 91 | "values = np.column_stack((scaled_features, scaled_label))\n", 92 | "\n", 93 | "n_train_hours = 365 * 24 *4\n", 94 | "train = values[:n_train_hours, :]\n", 95 | "test = values[n_train_hours:, :]\n", 96 | "train_X, train_y = train[:, :-1], train[:, -1]\n", 97 | "test_X, test_y = test[:, :-1], test[:, -1]\n" 98 | ] 99 | }, 100 | { 101 | "cell_type": "code", 102 | "execution_count": 5, 103 | "metadata": {}, 104 | "outputs": [ 105 | { 106 | "name": "stdout", 107 | "output_type": "stream", 108 | "text": [ 109 | "Epoch 1/50\n", 110 | "35040/35040 [==============================] - 3s 73us/sample - loss: 0.0259\n", 111 | "Epoch 2/50\n", 112 | "35040/35040 [==============================] - 2s 55us/sample - loss: 0.0141\n", 113 | "Epoch 3/50\n", 114 | "35040/35040 [==============================] - 2s 54us/sample - loss: 0.0142\n", 115 | "Epoch 4/50\n", 116 | "35040/35040 [==============================] - 2s 56us/sample - loss: 0.0141\n", 117 | "Epoch 5/50\n", 118 | "35040/35040 [==============================] - 2s 69us/sample - loss: 0.0142\n", 119 | "Epoch 6/50\n", 120 | "35040/35040 [==============================] - 2s 56us/sample - loss: 0.0142\n", 121 | "Epoch 7/50\n", 122 | "35040/35040 [==============================] - 2s 55us/sample - loss: 0.0142\n", 123 | "Epoch 8/50\n", 124 | "35040/35040 [==============================] - 2s 56us/sample - loss: 0.0140\n", 125 | "Epoch 9/50\n", 126 | "35040/35040 [==============================] - 2s 55us/sample - loss: 0.0140\n", 127 | "Epoch 10/50\n", 128 | "35040/35040 [==============================] - 2s 55us/sample - loss: 0.0141\n", 129 | "Epoch 11/50\n", 130 | "35040/35040 [==============================] - 2s 55us/sample - loss: 0.0141\n", 131 | "Epoch 12/50\n", 132 | "35040/35040 [==============================] - 2s 56us/sample - loss: 0.0140\n", 133 | "Epoch 13/50\n", 134 | "35040/35040 [==============================] - 2s 56us/sample - loss: 0.0140\n", 135 | "Epoch 14/50\n", 136 | "35040/35040 [==============================] - 2s 58us/sample - loss: 0.0141\n", 137 | "Epoch 15/50\n", 138 | "35040/35040 [==============================] - 2s 57us/sample - loss: 0.0139\n", 139 | "Epoch 16/50\n", 140 | "35040/35040 [==============================] - 2s 63us/sample - loss: 0.0140\n", 141 | "Epoch 17/50\n", 142 | "35040/35040 [==============================] - 2s 60us/sample - loss: 0.0140\n", 143 | "Epoch 18/50\n", 144 | "35040/35040 [==============================] - 2s 54us/sample - loss: 0.0139\n", 145 | "Epoch 19/50\n", 146 | "35040/35040 [==============================] - 2s 54us/sample - loss: 0.0140\n", 147 | "Epoch 20/50\n", 148 | "35040/35040 [==============================] - 2s 53us/sample - loss: 0.0140\n", 149 | "Epoch 21/50\n", 150 | "35040/35040 [==============================] - 2s 54us/sample - loss: 0.0138\n", 151 | "Epoch 22/50\n", 152 | "35040/35040 [==============================] - 2s 52us/sample - loss: 0.0140\n", 153 | "Epoch 23/50\n", 154 | "35040/35040 [==============================] - 2s 51us/sample - loss: 0.0139\n", 155 | "Epoch 24/50\n", 156 | "35040/35040 [==============================] - 2s 54us/sample - loss: 0.0139\n", 157 | "Epoch 25/50\n", 158 | "35040/35040 [==============================] - 2s 53us/sample - loss: 0.0139\n", 159 | "Epoch 26/50\n", 160 | "35040/35040 [==============================] - 2s 52us/sample - loss: 0.0139\n", 161 | "Epoch 27/50\n", 162 | "35040/35040 [==============================] - 2s 52us/sample - loss: 0.0139\n", 163 | "Epoch 28/50\n", 164 | "35040/35040 [==============================] - 2s 53us/sample - loss: 0.0139\n", 165 | "Epoch 29/50\n", 166 | "35040/35040 [==============================] - 2s 51us/sample - loss: 0.0138\n", 167 | "Epoch 30/50\n", 168 | "35040/35040 [==============================] - 2s 53us/sample - loss: 0.0139\n", 169 | "Epoch 31/50\n", 170 | "35040/35040 [==============================] - 3s 82us/sample - loss: 0.0139\n", 171 | "Epoch 32/50\n", 172 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0138\n", 173 | "Epoch 33/50\n", 174 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0138\n", 175 | "Epoch 34/50\n", 176 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0139\n", 177 | "Epoch 35/50\n", 178 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0138\n", 179 | "Epoch 36/50\n", 180 | "35040/35040 [==============================] - 4s 103us/sample - loss: 0.0137\n", 181 | "Epoch 37/50\n", 182 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0139\n", 183 | "Epoch 38/50\n", 184 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0137\n", 185 | "Epoch 39/50\n", 186 | "35040/35040 [==============================] - 3s 97us/sample - loss: 0.0138\n", 187 | "Epoch 40/50\n", 188 | "35040/35040 [==============================] - 3s 97us/sample - loss: 0.0137\n", 189 | "Epoch 41/50\n", 190 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0138\n", 191 | "Epoch 42/50\n", 192 | "35040/35040 [==============================] - 3s 98us/sample - loss: 0.0138\n", 193 | "Epoch 43/50\n", 194 | "35040/35040 [==============================] - 3s 97us/sample - loss: 0.0137\n", 195 | "Epoch 44/50\n", 196 | "35040/35040 [==============================] - 3s 96us/sample - loss: 0.0138\n", 197 | "Epoch 45/50\n", 198 | "35040/35040 [==============================] - 3s 96us/sample - loss: 0.0137\n", 199 | "Epoch 46/50\n", 200 | "35040/35040 [==============================] - 3s 96us/sample - loss: 0.0138\n", 201 | "Epoch 47/50\n", 202 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0138\n", 203 | "Epoch 48/50\n", 204 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0137\n", 205 | "Epoch 49/50\n", 206 | "35040/35040 [==============================] - 3s 95us/sample - loss: 0.0137\n", 207 | "Epoch 50/50\n", 208 | "35040/35040 [==============================] - 3s 96us/sample - loss: 0.0137\n", 209 | "This took 126.04263424873352 seconds.\n" 210 | ] 211 | } 212 | ], 213 | "source": [ 214 | "model = tf.keras.Sequential([keras.layers.Dense(units=10, input_shape=[8]),keras.layers.Dense(units=5, input_shape=[10]), keras.layers.Dense(units=1, input_shape=[5])])\n", 215 | "\n", 216 | "model.compile(optimizer='adam', loss='mae')\n", 217 | "\n", 218 | "start = time.time()\n", 219 | "\n", 220 | "\n", 221 | "model.fit(train_X, train_y, epochs=50)\n", 222 | "end = time.time()\n", 223 | "print('This took {} seconds.'.format(end - start))\n", 224 | "\n", 225 | "y_pred = model.predict(test_X)" 226 | ] 227 | }, 228 | { 229 | "cell_type": "code", 230 | "execution_count": 4, 231 | "metadata": {}, 232 | "outputs": [ 233 | { 234 | "data": { 235 | "image/png": "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\n", 236 | "text/plain": [ 237 | "
" 238 | ] 239 | }, 240 | "metadata": { 241 | "needs_background": "light" 242 | }, 243 | "output_type": "display_data" 244 | }, 245 | { 246 | "name": "stdout", 247 | "output_type": "stream", 248 | "text": [ 249 | "Root Mean Squared Error: 0.0245\n", 250 | "R2 score : 0.93\n" 251 | ] 252 | } 253 | ], 254 | "source": [ 255 | "def plot_predicted(predicted_data, true_data):\n", 256 | " fig, ax = plt.subplots(figsize=(17,8))\n", 257 | " ax.set_title('Prediction vs. Actual after 50 epochs of training')\n", 258 | " ax.plot(true_data, label='True Data', color='green', linewidth='3')\n", 259 | "\n", 260 | " ax.plot(predicted_data, label='Prediction', color='red', linewidth='2')\n", 261 | " plt.legend()\n", 262 | " plt.show()\n", 263 | " \n", 264 | "rmse = np.sqrt(mean_squared_error(test_y, y_pred))\n", 265 | "test_y = scaler.inverse_transform(test_y.reshape(-1,1))\n", 266 | "y_pred = scaler.inverse_transform(y_pred.reshape(-1,1))\n", 267 | " \n", 268 | " \n", 269 | "plot_predicted(y_pred[:300,], test_y[:300,])\n", 270 | "print('Root Mean Squared Error: {:.4f}'.format(rmse))\n", 271 | "print(\"R2 score : %.2f\" % r2_score(test_y,y_pred))\n" 272 | ] 273 | }, 274 | { 275 | "cell_type": "code", 276 | "execution_count": null, 277 | "metadata": {}, 278 | "outputs": [], 279 | "source": [] 280 | } 281 | ], 282 | "metadata": { 283 | "kernelspec": { 284 | "display_name": "Python 3", 285 | "language": "python", 286 | "name": "python3" 287 | }, 288 | "language_info": { 289 | "codemirror_mode": { 290 | "name": "ipython", 291 | "version": 3 292 | }, 293 | "file_extension": ".py", 294 | "mimetype": "text/x-python", 295 | "name": "python", 296 | "nbconvert_exporter": "python", 297 | "pygments_lexer": "ipython3", 298 | "version": "3.7.2" 299 | } 300 | }, 301 | "nbformat": 4, 302 | "nbformat_minor": 2 303 | } 304 | -------------------------------------------------------------------------------- /algos.py: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env python3 2 | # -*- coding: utf-8 -*- 3 | """ 4 | Created on Sun Mar 3 23:19:01 2019 5 | 6 | @author: retroflake 7 | """ 8 | 9 | import pandas as pd 10 | import numpy as np 11 | import matplotlib.pyplot as plt 12 | plt.style.use('ggplot') 13 | plt.rcParams["figure.figsize"] = [16,9] 14 | 15 | #reading from the dataframe from the pickle file 16 | rf=pd.read_pickle('dframe.pkl') 17 | aqi =pd.read_pickle('final_aqi.pkl') 18 | aqi = np.asarray(aqi) 19 | 20 | #feature scaling 21 | from sklearn.preprocessing import StandardScaler 22 | scaler=StandardScaler() 23 | print(scaler.fit(rf)) 24 | ds=scaler.transform(rf) 25 | 26 | #taking PM2.5 as the Dependent Variable 27 | y=aqi 28 | #Taking rest of the variables as features 29 | X=ds[:,6:] 30 | 31 | 32 | #splitting the variables into train and test variables 33 | from sklearn.model_selection import train_test_split 34 | 35 | X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=.2, random_state=1234) 36 | 37 | #converting the variables to integers(do not run this) 38 | X=X.astype(int) 39 | y=y.astype(int) 40 | 41 | #Linear Regression 42 | from sklearn.linear_model import LinearRegression 43 | reg=LinearRegression() 44 | reg.fit(X_train,y_train) 45 | y_pred=reg.predict(X_test) 46 | print(reg.score(X,y)) 47 | 48 | 49 | #calculating mean square error 50 | from sklearn.metrics import mean_squared_error 51 | mean_squared_error(y_test, y_pred) 52 | 53 | #SVM 54 | from sklearn.svm import LinearSVC 55 | from sklearn.linear_model import LogisticRegression 56 | from sklearn.metrics import classification_report,accuracy_score 57 | LSVC = LinearSVC() 58 | 59 | LSVC = LSVC.fit(X_train,y_train) 60 | y_pred = LSVC.predict(X_test) 61 | print(accuracy_score(y_test,y_pred)) 62 | print(classification_report(y_test,y_pred)) 63 | 64 | #KNN 65 | from sklearn.neighbors import KNeighborsClassifier 66 | from sklearn.metrics import accuracy_score 67 | KNN=KNeighborsClassifier(n_neighbors=5) 68 | KNN.fit(X_train,y_train) 69 | y_pred=(KNN.predict(X_test)) 70 | print(accuracy_score(y_test,y_pred)) 71 | print(classification_report(y_test,y_pred)) 72 | -------------------------------------------------------------------------------- /aqi.py: -------------------------------------------------------------------------------- 1 | epa={'O3':[[0,54],[55,70],[71,85],[86,105],[106,200],[201,250],[251,300]], 2 | 'O3hrly':[[],[],[125,164],[165,204],[205,404],[405,504],[505,604]], 3 | 'PM2.5':[[0.0,12.0],[12.1,35.4],[35.5,55.4],[55.5,150.4],[150.5,250.4],[250.5,350.4],[350.5,500.4]], 4 | 'PM10':[[0,54],[55,154],[155,254],[255,354],[355,424],[425,504],[505,604]], 5 | 'CO':[[0.0,4.4],[4.5,9.4],[9.5,12.4],[12.5,15.4],[15.5,30.4],[30.5,40.4],[40.5,50.4]], 6 | 'SO2':[[0,35],[36,75],[76,185],[186,304],[305,604],[605,804],[805,1004]], 7 | 'NO2':[[0,53],[54,100],[101,360],[361,649],[650,1249],[1250,1649],[1650,2049]], 8 | 'AQI':[[0,50],[51,100],[101,150],[151,200],[201,300],[301,400],[401,500]]} 9 | 10 | AQI_cat=['good','moderate','unhealthy for sensitive groups','unhealthy','very unhealthy', 'hazardous', 'hazardous'] 11 | 12 | def find_aqi(pollutant,conc): 13 | i=0 14 | if conc > epa[pollutant][5][1]: 15 | return int(epa['AQI'][5][1]) 16 | for range in epa[pollutant]: 17 | if conc>= range[0] and conc<=range[1]: 18 | break 19 | i+=1 20 | aqi = (float(epa['AQI'][i][1] - epa['AQI'][i][0])/(range[1]-range[0]))*(conc-range[0]) + epa['AQI'][i][0] 21 | return(int(aqi)) 22 | -------------------------------------------------------------------------------- /aqi.pyc: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/rishika028/Air-Quality-Index-prediction-using-ML-and-LSTM/dc3a28f7985da60ad854cc4d34489786a01cddeb/aqi.pyc -------------------------------------------------------------------------------- /aqivar.py: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env python3 2 | # -*- coding: utf-8 -*- 3 | """ 4 | Created on Sun Mar 3 21:10:53 2019 5 | 6 | @author: retroflake 7 | """ 8 | 9 | import aqi 10 | import pandas as pd 11 | import numpy as np 12 | 13 | #Reading the features from the pickle file 14 | df=pd.read_pickle('dframe.pkl') 15 | 16 | max_aqi = [] 17 | aqi_list = [] 18 | 19 | #finding the AQI from the given feature of pollutants 20 | for i in range(1012): 21 | max_aqi.append(aqi.find_aqi('PM2.5',int(df.iloc[i][0]))) 22 | max_aqi.append(aqi.find_aqi('PM10',int(df.iloc[i][1]))) 23 | max_aqi.append(aqi.find_aqi('O3',int(df.iloc[i][2]))) 24 | max_aqi.append(aqi.find_aqi('NO2',int(df.iloc[i][3]))) 25 | max_aqi.append(aqi.find_aqi('SO2',int(df.iloc[i][4]))) 26 | max_aqi.append(aqi.find_aqi('CO',int(df.iloc[i][5]))) 27 | aqi_list.append(max(max_aqi)) 28 | max_aqi = [] 29 | 30 | #Using pandas to convert the 'list' to a dataframe so that it can be saved in a pickle file 31 | aqi_list= pd.DataFrame(aqi_list) 32 | #Saving the dataframe in the pickel format 33 | aqi_list.to_pickle('final_aqi.pkl') 34 | 35 | #Just an example to show how to import a pickle file and convert it into an array 36 | hh=pd.read_pickle('final_aqi.pkl') 37 | hh= np.asarray(hh) 38 | 39 | -------------------------------------------------------------------------------- /dframe.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/rishika028/Air-Quality-Index-prediction-using-ML-and-LSTM/dc3a28f7985da60ad854cc4d34489786a01cddeb/dframe.pkl -------------------------------------------------------------------------------- /final report.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/rishika028/Air-Quality-Index-prediction-using-ML-and-LSTM/dc3a28f7985da60ad854cc4d34489786a01cddeb/final report.pdf -------------------------------------------------------------------------------- /final_aqi.pkl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/rishika028/Air-Quality-Index-prediction-using-ML-and-LSTM/dc3a28f7985da60ad854cc4d34489786a01cddeb/final_aqi.pkl -------------------------------------------------------------------------------- /main.py: -------------------------------------------------------------------------------- 1 | import aqi 2 | aqi.find_aqi('PM10', 277) 3 | -------------------------------------------------------------------------------- /preprocessing.py: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env python3 2 | # -*- coding: utf-8 -*- 3 | """ 4 | Created on Sun Mar 3 23:40:30 2019 5 | 6 | @author: retroflake 7 | """ 8 | 9 | import pandas as pd 10 | import numpy as np 11 | import matplotlib.pyplot as plt 12 | 13 | 14 | plt.style.use('ggplot') 15 | plt.rcParams["figure.figsize"] = [16,9] 16 | df=pd.read_csv("../dataset/aqi.csv") 17 | 18 | df=df.drop(['locationid','stationname','chinesename','latitude','longitude','est_time'],axis=1) 19 | 20 | 21 | df['pm10']=df['pm10'].replace('-',np.NaN) 22 | df['o3']=df['o3'].replace('-',np.NaN) 23 | df['no2']=df['no2'].replace('-',np.NaN) 24 | df['so2']=df['so2'].replace('-',np.NaN) 25 | df['co']=df['co'].replace('-',np.NaN) 26 | df['temperature']=df['temperature'].replace('-',np.NaN) 27 | df['dewpoint']=df['dewpoint'].replace('-',np.NaN) 28 | df['pressure']=df['pressure'].replace('-',np.NaN) 29 | df['wind']=df['wind'].replace('-',np.NaN) 30 | df['humidity']=df['humidity'].replace('-',np.NaN) 31 | df=df.dropna() 32 | df=df.dropna(axis='columns') 33 | 34 | df 35 | df.to_pickle('dframe.pkl') -------------------------------------------------------------------------------- /svr.ipynb: -------------------------------------------------------------------------------- 1 | { 2 | "cells": [ 3 | { 4 | "cell_type": "code", 5 | "execution_count": 1, 6 | "metadata": {}, 7 | "outputs": [], 8 | "source": [ 9 | "from matplotlib import pyplot as plt\n", 10 | "from pandas import DataFrame\n", 11 | "from pandas import concat\n", 12 | "from sklearn.preprocessing import MinMaxScaler\n", 13 | "from sklearn.preprocessing import LabelEncoder\n", 14 | "from sklearn.metrics import mean_squared_error\n", 15 | "from sklearn.metrics import r2_score\n", 16 | "import numpy as np\n", 17 | "import pandas as pd\n", 18 | "\n", 19 | "def series_to_supervised(data, n_in=1, n_out=1, dropnan=True):\n", 20 | " n_vars = 1 if type(data) is list else data.shape[1]\n", 21 | " df = DataFrame(data)\n", 22 | " cols, names = list(), list()\n", 23 | " # input sequence (t-n, ... t-1)\n", 24 | " for i in range(n_in, 0, -1):\n", 25 | " cols.append(df.shift(i))\n", 26 | " names += [('var%d(t-%d)' % (j+1, i)) for j in range(n_vars)]\n", 27 | " # forecast sequence (t, t+1, ... t+n)\n", 28 | " for i in range(0, n_out):\n", 29 | " cols.append(df.shift(-i))\n", 30 | " if i == 0:\n", 31 | " names += [('var%d(t)' % (j+1)) for j in range(n_vars)]\n", 32 | " else:\n", 33 | " names += [('var%d(t+%d)' % (j+1, i)) for j in range(n_vars)]\n", 34 | " # put it all together\n", 35 | " agg = concat(cols, axis=1)\n", 36 | " agg.columns = names\n", 37 | " # drop rows with NaN values\n", 38 | " if dropnan:\n", 39 | " agg.dropna(inplace=True)\n", 40 | " return agg" 41 | ] 42 | }, 43 | { 44 | "cell_type": "code", 45 | "execution_count": 7, 46 | "metadata": {}, 47 | "outputs": [ 48 | { 49 | "name": "stdout", 50 | "output_type": "stream", 51 | "text": [ 52 | " var1(t-1) var2(t-1) var3(t-1) var4(t-1) var5(t-1) var6(t-1) \\\n", 53 | "1 129.0 -16.0 -4.0 1020.0 2.0 1.79 \n", 54 | "2 148.0 -15.0 -4.0 1020.0 2.0 2.68 \n", 55 | "3 159.0 -11.0 -5.0 1021.0 2.0 3.57 \n", 56 | "4 181.0 -7.0 -5.0 1022.0 2.0 5.36 \n", 57 | "5 138.0 -7.0 -5.0 1022.0 2.0 6.25 \n", 58 | "\n", 59 | " var7(t-1) var8(t-1) var1(t) \n", 60 | "1 0.0 0.0 148.0 \n", 61 | "2 0.0 0.0 159.0 \n", 62 | "3 0.0 0.0 181.0 \n", 63 | "4 1.0 0.0 138.0 \n", 64 | "5 2.0 0.0 109.0 \n" 65 | ] 66 | } 67 | ], 68 | "source": [ 69 | "dataset = pd.read_csv('pollution.csv', header=0, index_col=0)\n", 70 | "values = dataset.values\n", 71 | "# integer encode direction\n", 72 | "encoder = LabelEncoder()\n", 73 | "values[:,4] = encoder.fit_transform(values[:,4])\n", 74 | "# ensure all data is float\n", 75 | "values = values.astype('float32')\n", 76 | "reframed = series_to_supervised(values, 1, 1)\n", 77 | "reframed.drop(reframed.columns[[-7,-6,-5,-4,-3,-2,-1]], axis=1, inplace=True)\n", 78 | "print(reframed.head())" 79 | ] 80 | }, 81 | { 82 | "cell_type": "code", 83 | "execution_count": 8, 84 | "metadata": {}, 85 | "outputs": [], 86 | "source": [ 87 | "# split into train and test sets\n", 88 | "values = reframed.values\n", 89 | "scaler = MinMaxScaler(feature_range=(0, 1))\n", 90 | "scaled_features = scaler.fit_transform(values[:,:-1])\n", 91 | "scaled_label = scaler.fit_transform(values[:,-1].reshape(-1,1))\n", 92 | "values = np.column_stack((scaled_features, scaled_label))\n", 93 | "\n", 94 | "n_train_hours = 365 * 24 * 4\n", 95 | "train = values[:n_train_hours, :]\n", 96 | "test = values[n_train_hours:, :]\n", 97 | "# split into input and outputs\n", 98 | "# features take all values except the var1\n", 99 | "train_X, train_y = train[:, :-1], train[:, -1]\n", 100 | "test_X, test_y = test[:, :-1], test[:, -1]\n", 101 | "\n" 102 | ] 103 | }, 104 | { 105 | "cell_type": "code", 106 | "execution_count": 9, 107 | "metadata": {}, 108 | "outputs": [], 109 | "source": [ 110 | "from sklearn.svm import SVR\n", 111 | "\n", 112 | "x = train_X\n", 113 | "y = train_y\n", 114 | "\n", 115 | "regr = SVR(C = 2.0, epsilon = 0.1, kernel = 'rbf', gamma = 0.5, \n", 116 | " tol = 0.001, verbose=False, shrinking=True, max_iter = 10000)\n", 117 | "\n", 118 | "regr.fit(x, y)\n", 119 | "data_pred = regr.predict(x)\n", 120 | "y_pred = scaler.inverse_transform(data_pred.reshape(-1,1))\n", 121 | "y_inv = scaler.inverse_transform(y.reshape(-1,1))\n", 122 | "\n", 123 | "mse = mean_squared_error(y_inv, y_pred)\n", 124 | "rmse = np.sqrt(mse)\n", 125 | "\n", 126 | "def plot_predicted(predicted_data, true_data):\n", 127 | " fig, ax = plt.subplots(figsize=(17,8))\n", 128 | " ax.set_title('Prediction vs. Actual ')\n", 129 | " ax.plot(true_data, label='True Data', color='green', linewidth='3')\n", 130 | "\n", 131 | " ax.plot(predicted_data, label='Prediction', color='red', linewidth='2')\n", 132 | " plt.legend()\n", 133 | " plt.show()\n", 134 | " \n", 135 | "def run_test_nonlinear_reg(x, y):\n", 136 | " data_pred = regr.predict(x)\n", 137 | " y_pred = scaler.inverse_transform(data_pred.reshape(-1,1))\n", 138 | " y_inv = scaler.inverse_transform(y.reshape(-1,1))\n", 139 | "\n", 140 | " mse = mean_squared_error(y_inv, y_pred)\n", 141 | " rmse = np.sqrt(mse)\n", 142 | " print('Mean Squared Error: {:.4f}'.format(mse))\n", 143 | " print('Root Mean Squared Error: {:.4f}'.format(rmse))\n", 144 | "\n", 145 | " #Calculate R^2 (regression score function)\n", 146 | " print('Variance score: {:2f}'.format(r2_score(y_inv, y_pred)))\n", 147 | " return y_pred, y_inv" 148 | ] 149 | }, 150 | { 151 | "cell_type": "code", 152 | "execution_count": 11, 153 | "metadata": {}, 154 | "outputs": [ 155 | { 156 | "name": "stdout", 157 | "output_type": "stream", 158 | "text": [ 159 | "Mean Squared Error: 1776.6036\n", 160 | "Root Mean Squared Error: 42.1498\n", 161 | "Variance score: 0.797085\n" 162 | ] 163 | }, 164 | { 165 | "data": { 166 | "image/png": "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\n", 167 | "text/plain": [ 168 | "
" 169 | ] 170 | }, 171 | "metadata": { 172 | "needs_background": "light" 173 | }, 174 | "output_type": "display_data" 175 | }, 176 | { 177 | "name": "stdout", 178 | "output_type": "stream", 179 | "text": [ 180 | "Root Mean Squared Error: 45.4303\n", 181 | "R2 score : 0.80\n" 182 | ] 183 | } 184 | ], 185 | "source": [ 186 | "y_pred, y_inv = run_test_nonlinear_reg(test_X, test_y)\n", 187 | "plot_predicted(y_pred[:300,], y_inv[:300,])\n", 188 | "\n", 189 | "print('Root Mean Squared Error: {:.4f}'.format(rmse))\n", 190 | "\n", 191 | "print(\"R2 score : %.2f\" % r2_score(y_inv,y_pred))" 192 | ] 193 | }, 194 | { 195 | "cell_type": "code", 196 | "execution_count": null, 197 | "metadata": {}, 198 | "outputs": [], 199 | "source": [] 200 | } 201 | ], 202 | "metadata": { 203 | "kernelspec": { 204 | "display_name": "Python 3", 205 | "language": "python", 206 | "name": "python3" 207 | }, 208 | "language_info": { 209 | "codemirror_mode": { 210 | "name": "ipython", 211 | "version": 3 212 | }, 213 | "file_extension": ".py", 214 | "mimetype": "text/x-python", 215 | "name": "python", 216 | "nbconvert_exporter": "python", 217 | "pygments_lexer": "ipython3", 218 | "version": "3.7.2" 219 | } 220 | }, 221 | "nbformat": 4, 222 | "nbformat_minor": 2 223 | } 224 | --------------------------------------------------------------------------------