diff --git a/FINAL_LinearRegression.ipynb b/FINAL_LinearRegression.ipynb new file mode 100644 index 0000000..408431a --- /dev/null +++ b/FINAL_LinearRegression.ipynb @@ -0,0 +1,635 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "FINAL_LinearRegression.ipynb", + "provenance": [], + "collapsed_sections": [], + "toc_visible": true, + "authorship_tag": "ABX9TyPGqIkzfG1omzNb66FVl8+p", + "include_colab_link": true + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + } + }, + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zThzm0EfRCCm" + }, + "source": [ + "# Upload the MNIST datasets\n" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "0iK7yZKVP6mJ" + }, + "source": [ + "import numpy as np\n", + "import pandas as pd\n", + "import matplotlib.pyplot as plt" + ], + "execution_count": 29, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "dDZIAngEdVuA" + }, + "source": [ + "taking the mnist dataset in pandas dataframe" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "FBUH6qQCgzvu" + }, + "source": [ + "train_df = pd.read_csv(\"/content/sample_data/mnist_train_small.csv\")\n", + "test_df = pd.read_csv(\"/content/sample_data/mnist_test.csv\")" + ], + "execution_count": 6, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "42Re_lszhHU5" + }, + "source": [ + "# converting the panadas dataframe into numpy array\n", + "\n", + "train = train_df.to_numpy()\n", + "test = test_df.to_numpy()" + ], + "execution_count": 30, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "wflq-n7Xs_i0" + }, + "source": [ + "Splitting the dataset into 'labels' as 'y' and 'features' as 'x'" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "W8hPhLRMhoLn" + }, + "source": [ + "y_train = train[:,:1]\n", + "x_train = train[:,1:]\n", + "\n", + "y_test = test[:,:1]\n", + "x_test = test[:,1:]" + ], + "execution_count": 8, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "ekGsi8zhiqjK", + "outputId": "9cfaa968-ea1a-4256-aa2a-7b419b0e1fd8" + }, + "source": [ + "print(x_train.shape,y_train.shape)\n", + "print(x_test.shape,y_test.shape)" + ], + "execution_count": 9, + "outputs": [ + { + "output_type": "stream", + "text": [ + "(19999, 784) (19999, 1)\n", + "(9999, 784) (9999, 1)\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "IniJRQdOtNNL" + }, + "source": [ + "checking the number of training examples for each lable of digit" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "itFucbFJU5Bk", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "c72687a0-457f-4e45-b7d4-f6420707ef05" + }, + "source": [ + "for j in range(10):\n", + " temp = 0\n", + " for i in range(19999):\n", + " if y_train[i] == j:\n", + " temp = temp + 1\n", + " print('the',j,'digit occures',temp,'times')\n" + ], + "execution_count": 31, + "outputs": [ + { + "output_type": "stream", + "text": [ + "the 0 digit occures 1962 times\n", + "the 1 digit occures 2243 times\n", + "the 2 digit occures 1989 times\n", + "the 3 digit occures 2021 times\n", + "the 4 digit occures 1924 times\n", + "the 5 digit occures 1761 times\n", + "the 6 digit occures 2038 times\n", + "the 7 digit occures 2126 times\n", + "the 8 digit occures 1912 times\n", + "the 9 digit occures 2023 times\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "AmHTAtc_uAYf" + }, + "source": [ + "# Visualizing the dataset" + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 281 + }, + "id": "m4tX6ueRuFkp", + "outputId": "38625996-725b-4b99-82db-89a1f9bf999c" + }, + "source": [ + "# Reshape the array into 28 x 28 array (2-dimensional array)\n", + "index =12\n", + "pixels = x_train[index].reshape((28, 28))\n", + "\n", + "# Plot\n", + "plt.title('Label is'+str(y_train[index]))\n", + "plt.imshow(pixels, cmap='gray')\n", + "plt.show()" + ], + "execution_count": 37, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "UnT5CBwodh-T" + }, + "source": [ + "# Data Processing" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "PSzJnjqr7tJX" + }, + "source": [ + "# adding the '1' on the starting of each array rows\n", + "\n", + "x_n_train = np.zeros((19999,785))\n", + "\n", + "for i in range(19999):\n", + " x_n_train[i] = np.insert(x_train[i],0,1)\n", + "\n", + "x_n_train = x_n_train/255" + ], + "execution_count": 47, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "O6p3Eg8uBdpC", + "outputId": "673e1672-d20a-48ab-dd61-75170d27f15b" + }, + "source": [ + "print(x_n_train.shape)" + ], + "execution_count": 48, + "outputs": [ + { + "output_type": "stream", + "text": [ + "(19999, 785)\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "bON_aX29F8Y5" + }, + "source": [ + "# The Main Function" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "RYwWsN2GgwOA" + }, + "source": [ + "The Main Linear Regression Function Starts from here" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "TSPtq0DCzvTz" + }, + "source": [ + "''' the Q here is parameter here (THEETA)'''\n", + "\n", + "Q = np.random.randn(1,785)" + ], + "execution_count": 49, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "F1PIFIHwWrOv" + }, + "source": [ + "# definging some useful values\n", + "\n", + "itrations = 50000\n", + "alpha = 0.05\n", + "m =19999\n", + "cost_function = []" + ], + "execution_count": 50, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "aDjlc4eeCemM" + }, + "source": [ + "X = x_n_train.T\n", + "Y = y_train.T" + ], + "execution_count": 51, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "szjoDcUBRy6b", + "outputId": "ca39a416-372d-4205-9ee0-55b466fccf1e" + }, + "source": [ + "for i in range(1,itrations+1):\n", + "\n", + " #PREDICTIONS\n", + " hypothesis = np.dot(Q,X) \n", + "\n", + " # COST FUNCTION \n", + " j = 1/(2*m)*np.sum((hypothesis-Y)**2)\n", + " cost_function.append(j)\n", + "\n", + "\n", + " # GRADIENT DESCENT\n", + " dw = 1/m * np.dot(hypothesis-Y,X.T)\n", + " Q = Q - alpha*dw\n", + " \n", + "\n", + " # printing the cost function which shows the degressing nature \n", + " if i%5000 == 0:\n", + " print(i, 'th iteration, Cost Function =',j)\n", + "\n", + "\n", + " # logic to break the loop if the change in cost is very very low \n", + " # ie. which sigifies that our parameters are rechead to give the minimum lavel of cost function\n", + "\n", + " if i%2 == 0:\n", + " if abs(j-cost_function[-2])<0.000001:\n", + " if abs(j-cost_function[-3])<0.000001:\n", + " print('the last itration was ',i,'th','and the final cost function is ',j)\n", + " break " + ], + "execution_count": 52, + "outputs": [ + { + "output_type": "stream", + "text": [ + "5000 th iteration, Cost Function = 1.907491359455398\n", + "10000 th iteration, Cost Function = 1.8530785689178375\n", + "15000 th iteration, Cost Function = 1.837599896804272\n", + "20000 th iteration, Cost Function = 1.8300714061720498\n", + "25000 th iteration, Cost Function = 1.8254547906233693\n", + "30000 th iteration, Cost Function = 1.8222644844360654\n", + "the last itration was 31454 th and the final cost function is 1.8215070244278082\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "7lZZN-hpYejQ", + "outputId": "20e7faff-7356-4277-d20d-61065284c5e9" + }, + "source": [ + "# hypothesis = hypothesis.astype(int)\n", + "print(type(hypothesis),hypothesis.shape)\n", + "print(hypothesis)" + ], + "execution_count": 53, + "outputs": [ + { + "output_type": "stream", + "text": [ + " (1, 19999)\n", + "[[5.29224837 2.39935706 8.25798228 ... 3.2150219 6.9363307 4.89634861]]\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fiVSqY2lhQYf" + }, + "source": [ + "Graph between Cost function vs no. of iteration" + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 265 + }, + "id": "549HqnK5Ww_D", + "outputId": "dbcc0352-993b-48a2-ffad-c183ab60d271" + }, + "source": [ + "# this graph shows the decreasing nature of cost function\n", + "\n", + "plt.plot(cost_function)\n", + "plt.show()" + ], + "execution_count": 54, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "QCLPNinWc5HO" + }, + "source": [ + "# Testing" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "IRPrJGBUMZbQ" + }, + "source": [ + "y_pred = hypothesis.round()\n", + "y_n_pred = y_pred.astype(int)\n", + "\n", + "for i in range(19999):\n", + " if y_n_pred[0][i] < 0 :\n", + " y_n_pred[0][i] = 0\n", + " elif y_n_pred[0][i] >=10:\n", + " y_n_pred[0][i] = 9" + ], + "execution_count": 55, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "K0ZCP8WFPtt1", + "outputId": "9d2bcc6d-22df-4bed-fe4e-b3d5f37cae6b" + }, + "source": [ + "k = 0\n", + "for i in range(19999):\n", + " # print(hypothesis[0][i],y_train[i])\n", + " if y_pred[0][i] == y_train[i]:\n", + " k = k+1\n", + "print(k,'is the number of corect prediction','the accurecy is ',k/199.99,'%')" + ], + "execution_count": 58, + "outputs": [ + { + "output_type": "stream", + "text": [ + "4961 is the number of corect prediction the accurecy is 24.8062403120156 %\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "mh3KqIKTy7PM" + }, + "source": [ + "Testing on the defined testing labels" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "JvZ9RNFulk4r" + }, + "source": [ + "x_n_test = np.random.randn(9999,785)\n", + "\n", + "for i in range(9999):\n", + " x_n_test[i] = np.insert(x_train[i],0,1)\n", + "\n", + "x_n_test = x_n_test/255" + ], + "execution_count": 59, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "Upfm_KY_kz16" + }, + "source": [ + "y_test_pred = np.dot(Q,x_n_test.T)" + ], + "execution_count": 60, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "Q0TmMY4ol7b7" + }, + "source": [ + "y_test_pred = y_test_pred.astype(int)" + ], + "execution_count": 61, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "REatbBj9mZFn", + "outputId": "b50500ec-f819-4a39-acb0-318176537447" + }, + "source": [ + "no_of_correct_predictions = 0\n", + "for i in range(9999):\n", + " if y_test_pred[0][i] == y_test[i]:\n", + " no_of_correct_predictions += 1\n", + "\n", + "print(no_of_correct_predictions,'is the number of corect prediction','the accurecy is ',no_of_correct_predictions/99.99,'%')" + ], + "execution_count": 62, + "outputs": [ + { + "output_type": "stream", + "text": [ + "932 is the number of corect prediction the accurecy is 9.320932093209322 %\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "EqV1ZUvRoxIc" + }, + "source": [ + "# Accuracy Using Sklearn\n" + ] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "teCeL0M-owHP", + "outputId": "d6bc76cf-e581-43a0-97cf-62dd6158fe06" + }, + "source": [ + "from sklearn.linear_model import LinearRegression\n", + "from sklearn.metrics import accuracy_score\n", + "\n", + "reg = LinearRegression()\n", + "reg.fit(x_train,y_train)\n", + "\n", + "y_pred = reg.predict(x_test)\n", + "\n", + "y_pred = y_pred.astype(int)\n", + "\n", + "accuracy = accuracy_score(y_pred,y_test)\n", + "\n", + "print('accuracy of the model using sklearn is',accuracy*100,'%')" + ], + "execution_count": 63, + "outputs": [ + { + "output_type": "stream", + "text": [ + "accuracy of the model using sklearn is 25.16251625162516 %\n" + ], + "name": "stdout" + } + ] + } + ] +} \ No newline at end of file diff --git a/MNIST_LinearRegression_final.ipynb b/MNIST_LinearRegression_final.ipynb new file mode 100644 index 0000000..5355819 --- /dev/null +++ b/MNIST_LinearRegression_final.ipynb @@ -0,0 +1,235 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "MNIST_LinearRegression_final.ipynb", + "provenance": [], + "authorship_tag": "ABX9TyN3GMRtSCCsCFDbuiXQ19C2", + "include_colab_link": true + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + } + }, + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "zThzm0EfRCCm" + }, + "source": [ + "# Loading MNIST data" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "0iK7yZKVP6mJ" + }, + "source": [ + "import numpy as np\r\n", + "import sklearn\r\n", + "import matplotlib.pyplot as plt" + ], + "execution_count": 2, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "E1c-cIo6QOgn" + }, + "source": [ + "from sklearn.datasets import fetch_openml\r\n", + "mnist = fetch_openml('mnist_784')" + ], + "execution_count": 4, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "4GC1j3AMQe5d", + "outputId": "8d60a1a4-fcf1-4f2c-f064-873982246a3c" + }, + "source": [ + "x = mnist['data']\r\n", + "y = mnist['target']\r\n", + "print(x.shape)\r\n", + "print(y.shape)" + ], + "execution_count": 5, + "outputs": [ + { + "output_type": "stream", + "text": [ + "(70000, 784)\n", + "(70000,)\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "kJWWZH7cQ14M" + }, + "source": [ + "x_train = x[:600]\r\n", + "# x_test = x[60000:]\r\n", + "y_train = y[:600]\r\n", + "# y_test = y[60000:]" + ], + "execution_count": 31, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "y51cASy8Q6np" + }, + "source": [ + "y_train = y_train.astype(np.int8)\r\n", + "y_test = y_test.astype(np.int8)" + ], + "execution_count": 32, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fvxWU-VrRJip" + }, + "source": [ + "# Training" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "OMb5lt-lQ9mZ" + }, + "source": [ + "def mult_vector(A,B):\r\n", + " for i in range(785):\r\n", + " r = A[i]*B[i] \r\n", + " \r\n", + " return r" + ], + "execution_count": 33, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "TqTKo9GMRfmc", + "outputId": "4ca8ede4-25a7-4803-e396-84293c24c81b" + }, + "source": [ + "Q = np.array([2]*785)\r\n", + "print(Q.shape)\r\n", + "# print(Q)" + ], + "execution_count": 34, + "outputs": [ + { + "output_type": "stream", + "text": [ + "(785,)\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "szjoDcUBRy6b" + }, + "source": [ + "def main_function(Q,x_train,y_train,alpha,no_of_itration):\r\n", + " m = len(x_train)\r\n", + "\r\n", + " y_pred = np.array([1]*600)\r\n", + " # xn = x_train\r\n", + " J = 1\r\n", + " for j in range(no_of_itration):\r\n", + "\r\n", + " for i in range(600):\r\n", + " xn = np.insert(x_train[i],0,1)\r\n", + " y_pred[i] = mult_vector(Q,xn)\r\n", + " \r\n", + " x_n_train = np.transpose(xn)\r\n", + " J = np.sum((y_pred - y_train)**2/(2*m))\r\n", + "\r\n", + " for i in range(785):\r\n", + " Q[i] = Q[i] - alpha*(1/m)*np.sum(x_n_train[i]*(y_pred - y_train))\r\n", + " \r\n", + " if j%5==0:\r\n", + " print(J)\r\n", + "\r\n", + " return Q\r\n" + ], + "execution_count": 53, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "BU1v5FYRWfCL", + "outputId": "d930df52-6297-4de5-ac65-504be889cb4a" + }, + "source": [ + "Q = main_function(Q,x_train,y_train,0.1,26)\r\n" + ], + "execution_count": 54, + "outputs": [ + { + "output_type": "stream", + "text": [ + "13.723333333333333\n", + "13.723333333333333\n", + "13.723333333333333\n", + "13.723333333333333\n", + "13.723333333333333\n", + "13.723333333333333\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "549HqnK5Ww_D" + }, + "source": [ + "" + ], + "execution_count": null, + "outputs": [] + } + ] +} \ No newline at end of file diff --git a/MNIST_Original.ipynb b/MNIST_Original.ipynb new file mode 100644 index 0000000..8a84fe1 --- /dev/null +++ b/MNIST_Original.ipynb @@ -0,0 +1,286 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "MNIST_Original.ipynb", + "provenance": [], + "collapsed_sections": [], + "toc_visible": true, + "authorship_tag": "ABX9TyOpsDmv8BaCz3mWNjobH0LE", + "include_colab_link": true + }, + "kernelspec": { + "name": "python3", + "display_name": "Python 3" + } + }, + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "WpGEnfGcSFK3" + }, + "source": [ + "# Importing mnist data\r\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "0eelf_9qDP7V" + }, + "source": [ + "**Importing MNIST data**" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "rf3VgdgFSOSd" + }, + "source": [ + "import numpy as np\r\n", + "import pandas as pd\r\n", + "import sklearn\r\n", + "import matplotlib.pyplot as plt" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "W6musYZwslmT" + }, + "source": [ + "from sklearn.datasets import fetch_openml\r\n", + "mnist = fetch_openml('mnist_784')" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "LrtvcPQN4QQl" + }, + "source": [ + "# from keras.datasets import mnist\r\n", + "# (x_train, y_train), (x_test, y_test) = mnist.load_data()" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "w0C2Ooo1t79b" + }, + "source": [ + "x = mnist['data']\r\n", + "y = mnist['target']" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "M3cBTCWv43NG" + }, + "source": [ + "# print(x.shape)\r\n", + "# print(type(x))" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "DZxfhvCOwQEm" + }, + "source": [ + "x_train = x[:60000]\r\n", + "x_test = x[60000:]\r\n", + "y_train = y[:60000]\r\n", + "y_test = y[60000:]" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "wSCL4sYoxUus" + }, + "source": [ + "y_train = y_train.astype(np.int8)\r\n", + "y_test = y_test.astype(np.int8)\r\n", + "\r\n", + "# print(y_train[2343])\r\n", + "# print(x_train[1].shape)" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "g2JIlQyiwJ1w" + }, + "source": [ + "visualization" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "iAhGNu1H9En9" + }, + "source": [ + "# digit = x_train[30040]\r\n", + "# plt.imshow(digit, cmap=matplotlib.cm.binary, interpolation=\"nearest\")\r\n", + "# #plt.axis('off')\r\n", + "# y_train[30040]" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "0OR1TSoqlGH9" + }, + "source": [ + "# applying Linear Regression" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "L2aRLdharXf0" + }, + "source": [ + "Defining Hypothesis" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "Z_sht0YUCXJ0" + }, + "source": [ + "def mult_vector(A,B):\r\n", + " for i in range(784):\r\n", + " r = A[i]*B[i] \r\n", + " \r\n", + " return r" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "at7Vizzk1qei" + }, + "source": [ + "from numpy import random" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "T0a-MraSzqhM" + }, + "source": [ + "Q = random.randint(50, size=(784))\r\n", + "X = random.randint(50, size=(784))\r\n", + "H = mult_vector(Q,X)" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "F3i_8Eqn-K9e" + }, + "source": [ + "Transpose Of Matrix" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "YagHiKtR-PjX" + }, + "source": [ + "def transpose(A,B): \r\n", + " \r\n", + " for i in range(N): \r\n", + " for j in range(N): \r\n", + " B[i][j] = A[j][i] \r\n", + " return B" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "GGfnT1Jq0uvs" + }, + "source": [ + "def main_function(Q,X,alpha,no_of_itration):\r\n", + " x_new_train = random.randint(50, size=(784))\r\n", + " x_n_train = transpose(x_train,x_new_train)\r\n", + " for j in range(no_of_itration):\r\n", + "\r\n", + " for i in range(784):\r\n", + " Q[i] = Q[i] - alpha*np.sum((H - y_train)*x_n_train**i)/784\r\n", + "\r\n", + " return Q" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "hPNQaxbD2Qun" + }, + "source": [ + "Q = main_function(Q,x_train,0.001,100)\r\n", + "y_pred_test = mult_vector(Q,x_test)" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "kvcLtIZN5zj0" + }, + "source": [ + "" + ], + "execution_count": null, + "outputs": [] + } + ] +} \ No newline at end of file diff --git a/README.md b/README.md index 2a85d2a..f9852d1 100644 --- a/README.md +++ b/README.md @@ -1,3 +1,6 @@ -# Logistic regression +Linear-Regression +# Linear Regression + +#this is for the implementation of linear regression on MNIST dataset + -this is for logistic regression implemntation on MNIST dataset. diff --git a/my_data.ipynb b/my_data.ipynb new file mode 100644 index 0000000..331d9b9 --- /dev/null +++ b/my_data.ipynb @@ -0,0 +1,478 @@ +{ + "nbformat": 4, + "nbformat_minor": 0, + "metadata": { + "colab": { + "name": "my_data.ipynb", + "provenance": [], + "collapsed_sections": [], + "toc_visible": true, + "authorship_tag": "ABX9TyO2k1GXflchAj6b548w37fX", + "include_colab_link": true + }, + "kernelspec": { + "display_name": "Python 3", + "name": "python3" + } + }, + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "view-in-github", + "colab_type": "text" + }, + "source": [ + "\"Open" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "R5-sOHH9p43a" + }, + "source": [ + "# Creating My Own Data\r\n" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "PXMUx7g1jTpW" + }, + "source": [ + "import pandas as pd\r\n", + "import numpy as np\r\n", + "import matplotlib\r\n", + "import matplotlib.pyplot as plt" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "9GyaxxwqFSfS" + }, + "source": [ + "# from google.colab import files\r\n", + "# uploded = files.upload()\r\n", + "# import io\r\n", + "# mera_data = pd.read_excel(io.BytesIO(uploded['test_data.xlsx']))\r\n", + "# mera_data" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "sSBMjCk_FUxi" + }, + "source": [ + "mera_data = np.array([[144, 300, 600, 1300, 1600, 2000, 3000, 3500, 4000, 5000],[400, 900, 2000, 5000, 9000, 14000, 16000, 18000, 20000, 25000]])" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "nIzZDUoDjDwP" + }, + "source": [ + "# print(type(mera_data))\r\n", + "# print(mera_data.shape)\r\n", + "\r\n", + "x_train=mera_data[0]\r\n", + "y_train=mera_data[1]\r\n", + "x_test = np.array([457,1000,1386,2456,3653,5786])\r\n", + "y_test = np.array([3*457,3*1000,3*1386,3*2456,3*3653,3*5786])" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "id": "ozXu9iFtjo_Z" + }, + "source": [ + "# plt.scatter(mera_data[0], mera_data[1])\r\n", + "# #plt.plot(mera_data[0], mera_data[1])\r\n", + "# plt.show()" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "n2Toa0qtKFEA" + }, + "source": [ + "\r\n", + "# Graph Ploting" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "UnTsqaxRKO82" + }, + "source": [ + "def plot(X, Y, X_test, Y_test):\r\n", + " plt.scatter( X, Y, c='r', marker='x' )\r\n", + " plt.plot(X_test, Y_test , '-o')\r\n", + " plt.show()" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 265 + }, + "id": "IQMrifwpKVUk", + "outputId": "00891dd6-b819-41cc-edd5-4f3444de9298" + }, + "source": [ + "plot(x_train, y_train, x_test, y_test)" + ], + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "THIVujJiqNI5" + }, + "source": [ + "# Finding the cofficient of best fited line\r\n" + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "o_ifRXkPqj9-" + }, + "source": [ + "Defining Cost function and Hypothesis" + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "i50FIz-YqjPb" + }, + "source": [ + "def linear_regression(no_of_itration,x,y,alpha):\r\n", + " Q_0 = 100\r\n", + " Q_1 = 4\r\n", + " \r\n", + " for i in range(no_of_itration):\r\n", + "\r\n", + " y_pred = Q_0 + Q_1*x\r\n", + "\r\n", + " cost_function = np.sum((y_pred - y)**2/(2*len(x)))\r\n", + "\r\n", + " grad_Q_1 = np.sum(x*(y_pred - y)/len(x))\r\n", + " grad_Q_0 = np.sum((y_pred-y)/len(x))\r\n", + "\r\n", + " Q_0 = Q_0 - alpha*grad_Q_0\r\n", + " Q_1 = Q_1 - alpha*grad_Q_1\r\n", + "\r\n", + " if i%10==0:\r\n", + " plot(x_train,y_train,x_train,y_pred)\r\n", + " print(cost_function)\r\n", + " list0 = [cost_function,Q_0,Q_1]\r\n", + " return list0" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "71czcddpEs8K", + "outputId": "c102e9ce-c072-46af-9408-24876e4dda12" + }, + "source": [ + "list0 = linear_regression(no_of_itration=100,x=x_train,y=y_train,alpha=0.0000001)\r\n", + "#print(list0)" + ], + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "5563808.8\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.4071850565\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.4011868031\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3952708011\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3893548037\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3834388104\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3775228213\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3716068354\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3656908553\n" + ], + "name": "stdout" + }, + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + }, + { + "output_type": "stream", + "text": [ + "1009677.3597748776\n" + ], + "name": "stdout" + } + ] + }, + { + "cell_type": "code", + "metadata": { + "id": "zccJh0jY6Ilf" + }, + "source": [ + "c = list0[1]\r\n", + "m = list0[2]" + ], + "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 265 + }, + "id": "sZnnyIMCgfOf", + "outputId": "4aa29176-8d1b-4133-aa20-7ccf4795deb0" + }, + "source": [ + "y_pred_test = [None]*len(x_test)\r\n", + "for i in range(6):\r\n", + " y_pred_test[i] = m*x_test[i] + c\r\n", + "\r\n", + "plt.scatter(x_test,y_test)\r\n", + "plt.plot(x_test,y_pred_test,'-o')\r\n", + "plt.show()" + ], + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "image/png": "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\n", + "text/plain": [ + "
" + ] + }, + "metadata": { + "tags": [], + "needs_background": "light" + } + } + ] + } + ] +} \ No newline at end of file