Check_predict_method.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Import Dependencies"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "from sklearn.datasets import load_digits\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "from mlp import Mlp\n",
    "import cv2 as cv\n",
    "from OneHotEncoder_OneHotDecoder_classes import OneHotEncoder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = load_digits()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((1437, 64), (1437, 10), (360, 64), (360, 10))"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = dataset.data\n",
    "Y = dataset.target\n",
    "\n",
    "encoder = OneHotEncoder()\n",
    "encoder.fit(Y)\n",
    "Y = np.array(encoder.transform(Y))\n",
    "\n",
    "x_train, x_test, y_train, y_test = train_test_split(X, Y, test_size=0.2)\n",
    "x_train.shape, y_train.shape, x_test.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Mlp(dimension_input=x_train.shape[1], dimension_output=y_train.shape[1], H1=128, H2=32,\n",
    "            epochs=100, lr=.001, act_func1='sigmoid', act_func2='sigmoid', act_func_output='softmax')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "  0%|          | 0/100 [00:00<?, ?it/s]"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:08<00:00, 12.00it/s]"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "loss_test: 0.015910050921797165, accuracy_test: 0.8888888888888888\n",
      "loss_train: 0.0024207044180093113, accuracy_train: 0.9916492693110647\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    }
   ],
   "source": [
    "losses, accuracies = model.fit(x_train, y_train)\n",
    "\n",
    "loss_train, accuracy_train = model.evaluate(x_train, y_train)\n",
    "loss_test, accuracy_test = model.evaluate(x_test, y_test)\n",
    "\n",
    "print(f\"loss_test: {loss_test}, accuracy_test: {accuracy_test}\")\n",
    "print(f\"loss_train: {loss_train}, accuracy_train: {accuracy_train}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.suptitle(\"Losses And Accuracies\")\n",
    "\n",
    "plt.subplot(1,2,1)\n",
    "plt.plot(losses)\n",
    "plt.ylabel(\"Loss\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "plt.plot(accuracies)\n",
    "plt.ylabel(\"Accuracy\")\n",
    "plt.xlabel(\"Epoch\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd4009cada0>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "image_0 = cv.imread('test_images/number_0.jpg')\n",
    "image_0 = cv.cvtColor(image_0 , cv.COLOR_BGR2GRAY)\n",
    "image_0 = cv.resize(image_0, (8, 8))\n",
    "plt.imshow(image_0, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = image_0.reshape(1, 64)\n",
    "y_pred = model.predict([x])\n",
    "np.argmax(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd4008c2860>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "image_2 = cv.imread('test_images/number_2.jpg')\n",
    "image_2 = cv.cvtColor(image_2 , cv.COLOR_BGR2GRAY)\n",
    "image_2 = cv.resize(image_2, (8, 8))\n",
    "plt.imshow(image_2, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/media/matin/New Volume/matin/python_work/pylearn/classes/ML/48/practices/mlp.py:30: RuntimeWarning: overflow encountered in exp\n",
      "  return 1 / (1 + np.exp(-x))\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = image_2.reshape(1, 64)\n",
    "y_pred = model.predict([x])\n",
    "np.argmax(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.image.AxesImage at 0x7fd40093b4c0>"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "image_9 = cv.imread('test_images/number_9.jpg')\n",
    "image_9 = cv.cvtColor(image_9 , cv.COLOR_BGR2GRAY)\n",
    "image_9 = cv.resize(image_9, (8, 8))\n",
    "plt.imshow(image_9, cmap='gray')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/media/matin/New Volume/matin/python_work/pylearn/classes/ML/48/practices/mlp.py:30: RuntimeWarning: overflow encountered in exp\n",
      "  return 1 / (1 + np.exp(-x))\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "9"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = image_9.reshape(1, 64)\n",
    "y_pred = model.predict([x])\n",
    "np.argmax(y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "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": "3.10.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}