Stochastic Gradient Desent to derive coefficients of non-linear models

SGD.ipynb

{"cells":[{"cell_type":"markdown","metadata":{"colab_type":"text","id":"JFdwu2REWOWj"},"source":["# Stochastic gradient descent to derive the coefficent updates for all 7 coefficients of the model: $$ y = w_0 + w_1*x_1 + w_2*x_1*x_2 + w_3*x_2 + w_4*x_2^2+w_5*x_1*x_2*x_3+w_6*x_3$$"]},{"cell_type":"code","execution_count":1,"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":435},"colab_type":"code","executionInfo":{"elapsed":443,"status":"error","timestamp":1587415683418,"user":{"displayName":"sik","photoUrl":"","userId":"00543779398285328827"},"user_tz":300},"id":"OSa9EocrWOWv","outputId":"0d5aeac3-a188-45e3-eda3-1a673cac3f4a"},"outputs":[{"ename":"FileNotFoundError","evalue":"ignored","output_type":"error","traceback":["\u001b[0;31m---------------------------------------------------------------------------\u001b[0m","\u001b[0;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)","\u001b[0;32m<ipython-input-1-f33c4022e7de>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      5\u001b[0m \u001b[0mget_ipython\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mmagic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'matplotlib inline'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 7\u001b[0;31m \u001b[0msample\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mpd\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mread_csv\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"samples1.csv\"\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mindex_col\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;36m0\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m","\u001b[0;32m/usr/local/lib/python3.6/dist-packages/pandas/io/parsers.py\u001b[0m in \u001b[0;36mparser_f\u001b[0;34m(filepath_or_buffer, sep, delimiter, header, names, index_col, usecols, squeeze, prefix, mangle_dupe_cols, dtype, engine, converters, true_values, false_values, skipinitialspace, skiprows, skipfooter, nrows, na_values, keep_default_na, na_filter, verbose, skip_blank_lines, parse_dates, infer_datetime_format, keep_date_col, date_parser, dayfirst, cache_dates, iterator, chunksize, compression, thousands, decimal, lineterminator, quotechar, quoting, doublequote, escapechar, comment, encoding, dialect, error_bad_lines, warn_bad_lines, delim_whitespace, low_memory, memory_map, float_precision)\u001b[0m\n\u001b[1;32m    674\u001b[0m         )\n\u001b[1;32m    675\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 676\u001b[0;31m         \u001b[0;32mreturn\u001b[0m \u001b[0m_read\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfilepath_or_buffer\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    677\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    678\u001b[0m     \u001b[0mparser_f\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m__name__\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mname\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.6/dist-packages/pandas/io/parsers.py\u001b[0m in \u001b[0;36m_read\u001b[0;34m(filepath_or_buffer, kwds)\u001b[0m\n\u001b[1;32m    446\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    447\u001b[0m     \u001b[0;31m# Create the parser.\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 448\u001b[0;31m     \u001b[0mparser\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mTextFileReader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mfp_or_buf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    449\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    450\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0mchunksize\u001b[0m \u001b[0;32mor\u001b[0m \u001b[0miterator\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.6/dist-packages/pandas/io/parsers.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, f, engine, **kwds)\u001b[0m\n\u001b[1;32m    878\u001b[0m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptions\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"has_index_names\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mkwds\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"has_index_names\"\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    879\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 880\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_make_engine\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mengine\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    881\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    882\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0mclose\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.6/dist-packages/pandas/io/parsers.py\u001b[0m in \u001b[0;36m_make_engine\u001b[0;34m(self, engine)\u001b[0m\n\u001b[1;32m   1112\u001b[0m     \u001b[0;32mdef\u001b[0m \u001b[0m_make_engine\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mengine\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"c\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1113\u001b[0m         \u001b[0;32mif\u001b[0m \u001b[0mengine\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"c\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1114\u001b[0;31m             \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_engine\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mCParserWrapper\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mf\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0moptions\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1115\u001b[0m         \u001b[0;32melse\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1116\u001b[0m             \u001b[0;32mif\u001b[0m \u001b[0mengine\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;34m\"python\"\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32m/usr/local/lib/python3.6/dist-packages/pandas/io/parsers.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, src, **kwds)\u001b[0m\n\u001b[1;32m   1889\u001b[0m         \u001b[0mkwds\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m\"usecols\"\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0musecols\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1890\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1891\u001b[0;31m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_reader\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mparsers\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mTextReader\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msrc\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m**\u001b[0m\u001b[0mkwds\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1892\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munnamed_cols\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0m_reader\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0munnamed_cols\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1893\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n","\u001b[0;32mpandas/_libs/parsers.pyx\u001b[0m in \u001b[0;36mpandas._libs.parsers.TextReader.__cinit__\u001b[0;34m()\u001b[0m\n","\u001b[0;32mpandas/_libs/parsers.pyx\u001b[0m in \u001b[0;36mpandas._libs.parsers.TextReader._setup_parser_source\u001b[0;34m()\u001b[0m\n","\u001b[0;31mFileNotFoundError\u001b[0m: [Errno 2] File samples1.csv does not exist: 'samples1.csv'"]}],"source":["import pandas as pd\n","import numpy as np\n","from numpy import random\n","import matplotlib.pyplot as plt\n","%matplotlib inline\n","\n","sample = pd.read_csv(\"samples1.csv\", index_col = 0)"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{},"colab_type":"code","id":"_KyQhyLyWOW0"},"outputs":[],"source":["#y = 0 + 0*sample.x_1 + 0*sample.x_1_x_2 + 0*sample.x_2 + 0*sample.x_2**2 + 0*sample.x_1_x_2_x_3 + 0*sample.x_3\n","y = sample.y.to_numpy()\n","X = sample.drop(\"y\", axis = 1).to_numpy()"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{},"colab_type":"code","id":"8KCXwSTLWOW4"},"outputs":[],"source":["class SGD():\n","    \"\"\"base class for SGD regression\"\"\"\n","    def __init__(self, learning_rate = 0.0 , num_epochs = 10, reg_constant = 0.0, batchsize = 1):\n","        self.learning_rate = learning_rate\n","        self.num_epochs = num_epochs\n","        self.reg_constant = reg_constant\n","        self.batch_size = int(batchsize)\n","        self.epoch_weights = [] #should be total 11 weights == one initial weights(0) + 10 updated weights after every epoch\n","        self.epoch_mses = []\n","        self.y_predicts = [] #[1 x 1000] for initial weights, and [1 x 1000] per epoch (11 in total)\n","        \n","    def iter_minibatches(self, X, y):\n","        shuffle=False #Although either way is fine, I avoid randomization, because randomization makes plots messier. \n","        assert X.shape[0] == y.shape[0]\n","        if shuffle:\n","            indices = np.arange(X.shape[0])\n","            np.random.shuffle(indices)\n","        for start_idx in range(0, X.shape[0] - self.batch_size + self.batch_size, self.batch_size):\n","            if shuffle:\n","                excerpt = indices[start_idx:start_idx + self.batch_size]\n","            else:\n","                excerpt = slice(start_idx, start_idx + self.batch_size)\n","            yield X[excerpt], y[excerpt]\n","    \n","    #define y predict\n","    def predict(self, X):\n","        \"\"\"y_predict\"\"\"\n","        #recall weights and create y_predicts list\n","        epoch_weights = self.epoch_weights #There are 11 weights: initial + 10 epochs = 11 in total\n","        epoch_weights = epoch_weights[1:11] #exculde initial weights because I don't have to calculate predicted y with initial weights (0)\n","        obj = {}\n","        for epoch in range(10): # Since I did not calculated predicted y with initial weights (0), there are ten predicted ys per epoch \n","            obj[epoch] = []\n","        for W_s, col in zip(epoch_weights, range(10)): #weight result per epoch\n","            for i in range(1000): #iterate through rows\n","                y_predict = W_s[0] + W_s[1]*X[i, 0] + W_s[2]*X[i, 0]*X[i, 1]  + W_s[3]*X[i, 1] + W_s[4]*X[i, 1]*X[i, 1] + W_s[5]*X[i,0]*X[i,1]*X[i,2] + W_s[6]*X[i,2]\n","                obj[col].append(y_predict)\n","                \n","        self.y_predicts = obj\n","        return obj     \n","     \n","    #define loss and iterate through mini batches\n","    def fit(self, X, y): \n","        \"\"\"Train model model\"\"\"\n","        epoch_weights = []\n","        W = np.array([0.0,0.0,0.0,0.0,0.0,0.0,0.0])\n","        epoch_weights.append(W) #append initial weights\n","        for epoch in range(self.num_epochs):\n","            single_error = []\n","            batches =  self.iter_minibatches(X, y)\n","            for batch_x, batch_y in batches:\n","                y_predict = W[0] + W[1]*batch_x[0,0] + W[2]*batch_x[0,0]*batch_x[0,1] + W[3]*batch_x[0,1] + W[4]*batch_x[0,1]*batch_x[0,1] + W[5]*batch_x[0,0]*batch_x[0,1]*batch_x[0,2] + W[6]*batch_x[0,2] \n","                \n","                #loss function\n","                loss = y_predict - batch_y\n","                error = np.sum(loss**2)\n","                single_error.append(error) \n","                \n","                #regularizer 1/2 * reg_constant * sum(weight**2)\n","                regular_derivative = self.reg_constant * np.sum(W)  \n","                \n","                partial_derivative = np.array([1,\n","                                               batch_x[0,0],\n","                                               batch_x[0,0]*batch_x[0,1],\n","                                               batch_x[0,1],\n","                                               batch_x[0,1]*batch_x[0,1],\n","                                               batch_x[0,0]*batch_x[0,1]*batch_x[0,2],\n","                                               batch_x[0,2]])\n","                gradient = 2 * loss * partial_derivative + regular_derivative\n","                W = W - (self.learning_rate * gradient)\n","            \n","            #append weights after iterating mini batches per each epoch\n","            epoch_weights.append(W)\n","            \n","            #append RMSE\n","            #mse = np.average(np.sum(single_error))\n","            #epoch_mses.append(mse)\n","            #print(F\"Epoch: {epoch + 1}, updated weights: {W}\")\n","            #print(F\"Epoch: {epoch + 1}, mse: {mse}\")\n","        self.epoch_weights = epoch_weights\n","        #self.epoch_mses = epoch_mses \n","            \n","    def weights(self):\n","        return self.epoch_weights\n","    \n","    def mses(self, y_pred, y_test):\n","        epoch_mses = []\n","        for epoch in range(10): \n","            #print(y_pred[epoch])\n","            mses_single_epoch = np.sum((y_pred[epoch] - y)**2)/len(y)\n","            epoch_mses.append(mses_single_epoch)\n","        self.epoch_mses = epoch_mses\n","        return epoch_mses       "]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{},"colab_type":"code","id":"9VwOAtTyWOW8","outputId":"8359a507-c074-4dc8-fb88-38ddb83e7a21"},"outputs":[{"name":"stdout","output_type":"stream","text":["learning rate: 1e-06, reg constant: 0\n","10 epochs MSEs: [322.5547185227467, 184.32095190678007, 122.92122011505653, 92.82970323483373, 76.28337195704985, 66.09953938911642, 59.20637820758299, 54.184860437502024, 50.317092178258335, 47.205574772470825]\n","learning rate: 1e-06, reg constant: 10\n","10 epochs MSEs: [325.47181853950326, 189.89921028434304, 129.62345946271054, 99.94150247696895, 83.69117264344983, 73.92178834349517, 67.59581563004615, 63.256809379775525, 60.13700804986565, 57.79742300105591]\n","learning rate: 1e-05, reg constant: 0\n","10 epochs MSEs: [51.25174438608913, 32.08295394524574, 22.75565890430717, 17.184615036778563, 13.769476729364689, 11.654760166964312, 10.334019648577355, 9.501304829179452, 8.970337038732556, 8.627063453520824]\n","learning rate: 1e-05, reg constant: 10\n","10 epochs MSEs: [62.956488178766186, 49.588862665909716, 42.77193529501842, 37.678417862623476, 33.65971722523029, 30.42963618934014, 27.807827828864216, 25.667423085089446, 23.91432782868972, 22.47627472698901]\n","learning rate: 0.0001, reg constant: 0\n","10 epochs MSEs: [12.052463716491639, 10.714397407328327, 10.419125372955317, 10.23604167717536, 10.103577321161099, 10.003973134377834, 9.927267546811017, 9.86696547556659, 9.818669994605024, 9.779346417954422]\n","learning rate: 0.0001, reg constant: 10\n","10 epochs MSEs: [26.79639015247241, 18.838761854470764, 17.923859824297356, 17.947602217796792, 18.025040127188596, 18.067614416393823, 18.086079835302222, 18.09298956804377, 18.0949175901365, 18.09481400088033]\n","learning rate: 0.001, reg constant: 0\n"]},{"name":"stderr","output_type":"stream","text":["C:\\Users\\js223\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:56: RuntimeWarning: overflow encountered in square\n","C:\\Users\\js223\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:69: RuntimeWarning: overflow encountered in multiply\n","C:\\Users\\js223\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:60: RuntimeWarning: invalid value encountered in double_scalars\n","C:\\Users\\js223\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:90: RuntimeWarning: overflow encountered in square\n"]},{"name":"stdout","output_type":"stream","text":["10 epochs MSEs: [4.3390740646060646e+147, 7.490518621986427e+292, inf, inf, nan, nan, nan, nan, nan, nan]\n","learning rate: 0.001, reg constant: 10\n","10 epochs MSEs: [1.234335246980924e+132, 1.7677699841407505e+261, inf, inf, nan, nan, nan, nan, nan, nan]\n","Best 2 combinations: lerning rate = 0.00001 & reg constant = 0, learning rate = 0.0001 & reg constant = 0\n"]},{"name":"stderr","output_type":"stream","text":["C:\\Users\\js223\\Anaconda3\\lib\\site-packages\\ipykernel_launcher.py:52: RuntimeWarning: invalid value encountered in double_scalars\n","C:\\Users\\js223\\Anaconda3\\lib\\site-packages\\numpy\\core\\fromnumeric.py:90: RuntimeWarning: invalid value encountered in reduce\n","  return ufunc.reduce(obj, axis, dtype, out, **passkwargs)\n"]}],"source":["# Fitting model / Predict ys == ys for 10 epochs / mses == calculate mse per each epoch / weights == initial weights + updated weights for 10 epochs\n","learning_rate = [0.000001, 0.00001, 0.0001, 0.001]\n","reg_constants = [0, 10]\n","num_epochs = 10\n","\n","for lr in learning_rate:\n","    for rc in reg_constants:\n","        model = SGD(learning_rate = lr, num_epochs = num_epochs, reg_constant = rc, batchsize = 1.0)\n","        print(f\"learning rate: {lr}, reg constant: {rc}\")\n","        model.fit(X,y)\n","        y_predict = model.predict(X) \n","        mses = model.mses(y_predict, y)\n","        \n","        #calculate the last epoch' rmse\n","        last_rmses = np.sqrt(mses[9])\n","        print(f\"10 epochs MSEs: {mses}\")\n","print(\"Best 2 combinations: lerning rate = 0.00001 & reg constant = 0, learning rate = 0.0001 & reg constant = 0\")"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{},"colab_type":"code","id":"LBNBvjxuWOXB","outputId":"091b9bbf-9c1d-4807-e3ff-d6f8e8ed78af"},"outputs":[{"name":"stdout","output_type":"stream","text":["Compared with MSEs using sklearn function, outputs from my mse function are the same.\n","Learning Rate: 1e-05, Reg Constant: 0, MSEs: [51.25174438608913, 32.08295394524574, 22.75565890430717, 17.184615036778563, 13.769476729364689, 11.654760166964312, 10.334019648577355, 9.501304829179452, 8.970337038732556, 8.627063453520824]\n","Learning Rate: 0.0001, Reg Constant: 0, MSEs: [12.052463716491639, 10.714397407328327, 10.419125372955317, 10.23604167717536, 10.103577321161099, 10.003973134377834, 9.927267546811017, 9.86696547556659, 9.818669994605024, 9.779346417954422]\n"]}],"source":["# Use Sklearn mean_squared_error function to verify if my mses are correct.\n","\n","from sklearn.metrics import mean_squared_error\n","learning_rate = [0.00001, 0.0001]\n","reg_constant = 0\n","num_epochs = 10\n","\n","print(\"Compared with MSEs using sklearn function, outputs from my mse function are the same.\")\n","\n","for lr in learning_rate:\n","        model = SGD(learning_rate = lr, num_epochs = num_epochs, reg_constant = reg_constant, batchsize = 1)\n","        mses_list = []\n","        model.fit(X,y)\n","        y_predict = model.predict(X)\n","        for i in range(10):\n","            mses = mean_squared_error(y, y_predict[i])\n","            mses_list.append(mses)\n","        \n","        print(f\"Learning Rate: {lr}, Reg Constant: {reg_constant}, MSEs: {mses_list}\")"]},{"cell_type":"code","execution_count":null,"metadata":{"colab":{},"colab_type":"code","id":"nWbbt-fWWOXK","outputId":"11da6956-e05c-49d6-eb0b-2e2a59a84ff3"},"outputs":[{"data":{"image/png":"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","text/plain":["<Figure size 1080x360 with 1 Axes>"]},"metadata":{"needs_background":"light","tags":[]},"output_type":"display_data"},{"data":{"image/png":"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","text/plain":["<Figure size 1080x360 with 1 Axes>"]},"metadata":{"needs_background":"light","tags":[]},"output_type":"display_data"},{"data":{"image/png":"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","text/plain":["<Figure size 1080x360 with 1 Axes>"]},"metadata":{"needs_background":"light","tags":[]},"output_type":"display_data"},{"data":{"image/png":"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","text/plain":["<Figure size 1080x360 with 1 Axes>"]},"metadata":{"needs_background":"light","tags":[]},"output_type":"display_data"}],"source":["#Plot MSE and the w parameters as a function of epoch count (10 epochs) for the best 2 combinations of learning_rate and regularization. ie you should have plots of MSE and parameter updates.\n","learning_rate = [0.00001, 0.0001]\n","reg_constant = 0\n","num_epochs = 10\n","\n","for lr in learning_rate:\n","    model = SGD(learning_rate = lr, num_epochs = num_epochs, reg_constant = reg_constant, batchsize = 1)\n","    model.fit(X,y)\n","    y_predict = model.predict(X)\n","    mses = model.mses(y_predict, y)\n","    #calculate the last epoch' rmse\n","    last_rmses = mses[9]\n","    #weights\n","    weights = model.weights()\n","    \n","    #plot -- MSE\n","    plt.figure(figsize=(15,5))\n","    plt.plot(np.arange(0, num_epochs), mses) # 10 mses = 10 epochs\n","    plt.title(f\"Learning Rate: {lr}, Regularization Constant: {reg_constant}, Final MSE after 10 epochs: {last_rmses}\", fontsize = 15)\n","    plt.xlabel(\"Epoch #\", fontsize = 15)\n","    plt.ylabel(\"MSE\", fontsize = 15)\n","    plt.show();\n","    #plot -- weights\n","    plt.figure(figsize=(15,5))\n","    plt.plot(np.arange(0, num_epochs + 1), weights) #11 weights in that it includes initial weights (0)\n","    plt.title(f\"Learning Rate: {lr}, Regularization Constant: {reg_constant}\", fontsize = 15)\n","    plt.legend([\"w0\", \"w1\", \"w2\", \"w3\", \"w4\", \"w5\", \"w6\"])\n","    plt.xlabel(\"Initial condition (0) + Epoch #\", fontsize = 15)\n","    plt.ylabel(\"Weights\", fontsize = 15)    \n","    plt.show();"]}],"metadata":{"colab":{"name":"HW3_v2_Sik-3.ipynb","provenance":[]},"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":"3.7.7"}},"nbformat":4,"nbformat_minor":0}