Skip to content

by Will Turbayevsky #149

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
turbaevsky wants to merge 11 commits into
base: master
Choose a base branch
from
Open

by Will Turbayevsky #149

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
11 commits
Select commit Hold shift + click to select a range
cb75f09
Create readme.md
turbaevsky Nov 23, 2019
9e7eadb
Add files via upload
turbaevsky Nov 23, 2019
eeadc74
Update readme.md
turbaevsky Nov 23, 2019
8d33c10
Create readme.md
turbaevsky Nov 23, 2019
4ce4962
Update readme.md
turbaevsky Nov 23, 2019
aa743ea
Add files via upload
turbaevsky Nov 23, 2019
447b054
Create readme.md
turbaevsky Nov 23, 2019
d7570ea
Add files via upload
turbaevsky Nov 23, 2019
6bf945e
Create readme.md
turbaevsky Nov 23, 2019
a7cd94a
Update readme.md
turbaevsky Nov 23, 2019
4f7f8a2
Add files via upload
turbaevsky Nov 23, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13,532 changes: 13,532 additions & 0 deletions challenge1/analysis/tur/ch1.html
View file
Open in desktop

Large diffs are not rendered by default.

362 changes: 362 additions & 0 deletions challenge1/analysis/tur/ch1.ipynb
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,362 @@
{
"cells": [
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
"import pandas as pd\n",
"import numpy as np\n",
"df = pd.read_csv('../../data/training_dataset_500.csv')"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {
"scrolled": true
},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>House</th>\n",
" <th>Year</th>\n",
" <th>Month</th>\n",
" <th>Temperature</th>\n",
" <th>Daylight</th>\n",
" <th>EnergyProduction</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>House</th>\n",
" <td>1.000000e+00</td>\n",
" <td>0.000000</td>\n",
" <td>-1.816873e-18</td>\n",
" <td>0.000881</td>\n",
" <td>0.001583</td>\n",
" <td>-0.008303</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Year</th>\n",
" <td>0.000000e+00</td>\n",
" <td>1.000000</td>\n",
" <td>-6.340757e-01</td>\n",
" <td>-0.356800</td>\n",
" <td>0.524603</td>\n",
" <td>0.267481</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Month</th>\n",
" <td>-1.816873e-18</td>\n",
" <td>-0.634076</td>\n",
" <td>1.000000e+00</td>\n",
" <td>0.353837</td>\n",
" <td>-0.276307</td>\n",
" <td>-0.232748</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Temperature</th>\n",
" <td>8.810764e-04</td>\n",
" <td>-0.356800</td>\n",
" <td>3.538369e-01</td>\n",
" <td>1.000000</td>\n",
" <td>-0.053363</td>\n",
" <td>0.272789</td>\n",
" </tr>\n",
" <tr>\n",
" <th>Daylight</th>\n",
" <td>1.582656e-03</td>\n",
" <td>0.524603</td>\n",
" <td>-2.763068e-01</td>\n",
" <td>-0.053363</td>\n",
" <td>1.000000</td>\n",
" <td>0.531577</td>\n",
" </tr>\n",
" <tr>\n",
" <th>EnergyProduction</th>\n",
" <td>-8.302696e-03</td>\n",
" <td>0.267481</td>\n",
" <td>-2.327484e-01</td>\n",
" <td>0.272789</td>\n",
" <td>0.531577</td>\n",
" <td>1.000000</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" House Year Month Temperature Daylight \\\n",
"House 1.000000e+00 0.000000 -1.816873e-18 0.000881 0.001583 \n",
"Year 0.000000e+00 1.000000 -6.340757e-01 -0.356800 0.524603 \n",
"Month -1.816873e-18 -0.634076 1.000000e+00 0.353837 -0.276307 \n",
"Temperature 8.810764e-04 -0.356800 3.538369e-01 1.000000 -0.053363 \n",
"Daylight 1.582656e-03 0.524603 -2.763068e-01 -0.053363 1.000000 \n",
"EnergyProduction -8.302696e-03 0.267481 -2.327484e-01 0.272789 0.531577 \n",
"\n",
" EnergyProduction \n",
"House -0.008303 \n",
"Year 0.267481 \n",
"Month -0.232748 \n",
"Temperature 0.272789 \n",
"Daylight 0.531577 \n",
"EnergyProduction 1.000000 "
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn.preprocessing import PolynomialFeatures, LabelEncoder, MinMaxScaler\n",
"df = df.drop(columns=['ID','Label'])\n",
"#df = df[df.Month.isin([4,5,6,7,8])]\n",
"df.corr()"
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [],
"source": [
"def preprocessing(df):\n",
" #from keras.utils import np_utils\n",
" X = df[['House','Month','Temperature','Daylight']]\n",
" #X = X[X.Month.isin([4,5,6,7,8])]\n",
" y = df[['EnergyProduction']]\n",
" #enc = LabelEncoder()\n",
" #house = X.House.values.reshape(-1,1)\n",
" #X['House'] = enc.fit(house).transform(house).toarray()\n",
" #sc = MinMaxScaler()\n",
" #X = sc.fit_transform(X)\n",
" return np.array(X), np.array(y).reshape(len(y),)\n",
" "
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"def MAPE(y_true, y_pred): \n",
" y_true, y_pred = np.array(y_true), np.array(y_pred)\n",
" return np.mean(np.abs((y_true - y_pred) / y_true))*100"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"LinearRegression(copy_X=True, fit_intercept=True, n_jobs=None, normalize=False)"
]
},
"execution_count": 22,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from sklearn import linear_model\n",
"from sklearn import svm\n",
"reg = linear_model.LinearRegression()\n",
"#reg = svm.SVR(kernel='rbf', C=8, gamma=5e-5)\n",
"X, y = preprocessing(df)\n",
"reg.fit(X, y)"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [],
"source": [
"dft = pd.read_csv('../../data/test_dataset_500.csv')\n",
"Xt, yt = preprocessing(dft)"
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"14.450698807918974"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"MAPE(yt, reg.predict(Xt))"
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [],
"source": [
"def individual(houseId, df=df):\n",
" X = df[['House','Month','Temperature','Daylight']]\n",
" X = X[X.House==houseId].drop(columns=['House'])\n",
" y = df[['House','EnergyProduction']]\n",
" y = y[y.House==houseId].drop(columns=['House'])\n",
" return np.array(X), np.array(y)"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {},
"outputs": [],
"source": [
"# individual prognoses\n",
"ireg = linear_model.LinearRegression()\n",
"y_pred, y_true = [],[]\n",
"with open('predicted_energy_production.csv','w') as f:\n",
" f.write('House, EnergyProduction\\n')\n",
" for i in set(dft.House.values):\n",
" x,y = individual(i)\n",
" ireg.fit(x,y)\n",
" p = ireg.predict(individual(i,dft)[0])[0][0]\n",
" y_pred.append(p)\n",
" y_true.append((individual(i,dft)[1])[0][0])\n",
" f.write('{}, {}\\n'.format(i,p))\n",
"with open('mape.txt','w') as f:\n",
" f.write(str(MAPE(y_true, y_pred)))\n"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"# neural network"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [],
"source": [
"from keras.models import Sequential\n",
"from keras.layers import Dense, Dropout, Activation"
]
},
{
"cell_type": "code",
"execution_count": 40,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Epoch 1/12\n",
"Epoch 2/12\n",
"Epoch 3/12\n",
"Epoch 4/12\n",
"Epoch 5/12\n",
"Epoch 6/12\n",
"Epoch 7/12\n",
"Epoch 8/12\n",
"Epoch 9/12\n",
"Epoch 10/12\n",
"Epoch 11/12\n",
"Epoch 12/12\n",
"500/500 [==============================] - 0s 138us/step\n"
]
},
{
"data": {
"text/plain": [
"[21690.6206796875, 17.908103942871094]"
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"model = Sequential()\n",
"model.add(Dense(units=40, activation='relu', input_dim=4))\n",
"model.add(Dense(units=20, activation='relu'))\n",
"model.add(Dense(1))\n",
"model.compile(optimizer='adam',\n",
" loss='mse',\n",
" metrics=['mape'])\n",
"\n",
"X, y = preprocessing(df)\n",
"model.fit(X, y, epochs=12, verbose=3)\n",
"\n",
"Xt, yt = preprocessing(dft)\n",
"model.evaluate(Xt, yt)"
]
},
{
"cell_type": "code",
"execution_count": 787,
"metadata": {},
"outputs": [],
"source": [
"#LTSE"
]
}
],
"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": "3.7.3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Binary file added challenge1/analysis/tur/ch1.pdf
View file
Open in desktop
Binary file not shown.
1 change: 1 addition & 0 deletions challenge1/analysis/tur/mape.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
7.955755908782016
Loading