{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "565d9ece-0afb-4bc9-ad00-46a61f701a45", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "2026-05-13 17:16:40.628343: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n", "2026-05-13 17:16:40.665518: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n", "To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n", "2026-05-13 17:16:41.558462: I tensorflow/core/util/port.cc:153] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\n" ] } ], "source": [ "import tensorflow as tf\n", "tf.config.set_visible_devices([], 'GPU')\n", "import pandas as pd" ] }, { "cell_type": "code", "execution_count": 2, "id": "a85329f7-1daa-40ca-b5de-50293bc4e2c1", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
letterx1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16
0I51237210554133928410
1D4116861062610373739
2N7116635946441061028
3G21311866665917510
4S41158388695660897
\n", "
" ], "text/plain": [ " letter x1 x2 x3 x4 x5 x6 x7 x8 x9 x10 x11 x12 x13 x14 x15 \\\n", "0 I 5 12 3 7 2 10 5 5 4 13 3 9 2 8 4 \n", "1 D 4 11 6 8 6 10 6 2 6 10 3 7 3 7 3 \n", "2 N 7 11 6 6 3 5 9 4 6 4 4 10 6 10 2 \n", "3 G 2 1 3 1 1 8 6 6 6 6 5 9 1 7 5 \n", "4 S 4 11 5 8 3 8 8 6 9 5 6 6 0 8 9 \n", "\n", " x16 \n", "0 10 \n", "1 9 \n", "2 8 \n", "3 10 \n", "4 7 " ] }, "execution_count": 2, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df = pd.read_csv('letter-recognition.data')\n", "\n", "cols = ['letter','x1','x2','x3','x4','x5','x6','x7','x8',\n", " 'x9','x10','x11','x12','x13','x14','x15','x16']\n", "cols_useful = ['x1','x2','x3','x4','x5','x6','x7','x8',\n", " 'x9','x10','x11','x12','x13','x14','x15','x16']\n", "\n", "df.columns = cols\n", "df.head()" ] }, { "cell_type": "code", "execution_count": 3, "id": "93b24cb3-9c84-402e-bc0c-7a019c3de325", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
x1x2x3x4x5x6x7x8x9x10x11x12x13x14x15x16
count19999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000019999.00000
mean4.0236517.0354525.1219565.3724693.5059756.8975457.5001754.6288315.1786098.2821646.4538237.9289963.0462528.3388673.6919357.80119
std1.9132063.3046312.0145682.2614452.1904412.0260712.3250872.6998372.3808752.4884852.6310162.0806712.3325001.5467592.5670041.61751
min0.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.0000000.00000
25%3.0000005.0000004.0000004.0000002.0000006.0000006.0000003.0000004.0000007.0000005.0000007.0000001.0000008.0000002.0000007.00000
50%4.0000007.0000005.0000006.0000003.0000007.0000007.0000004.0000005.0000008.0000006.0000008.0000003.0000008.0000003.0000008.00000
75%5.0000009.0000006.0000007.0000005.0000008.0000009.0000006.0000007.00000010.0000008.0000009.0000004.0000009.0000005.0000009.00000
max15.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000015.00000
\n", "
" ], "text/plain": [ " x1 x2 x3 x4 x5 \\\n", "count 19999.000000 19999.000000 19999.000000 19999.000000 19999.000000 \n", "mean 4.023651 7.035452 5.121956 5.372469 3.505975 \n", "std 1.913206 3.304631 2.014568 2.261445 2.190441 \n", "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n", "25% 3.000000 5.000000 4.000000 4.000000 2.000000 \n", "50% 4.000000 7.000000 5.000000 6.000000 3.000000 \n", "75% 5.000000 9.000000 6.000000 7.000000 5.000000 \n", "max 15.000000 15.000000 15.000000 15.000000 15.000000 \n", "\n", " x6 x7 x8 x9 x10 \\\n", "count 19999.000000 19999.000000 19999.000000 19999.000000 19999.000000 \n", "mean 6.897545 7.500175 4.628831 5.178609 8.282164 \n", "std 2.026071 2.325087 2.699837 2.380875 2.488485 \n", "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n", "25% 6.000000 6.000000 3.000000 4.000000 7.000000 \n", "50% 7.000000 7.000000 4.000000 5.000000 8.000000 \n", "75% 8.000000 9.000000 6.000000 7.000000 10.000000 \n", "max 15.000000 15.000000 15.000000 15.000000 15.000000 \n", "\n", " x11 x12 x13 x14 x15 \\\n", "count 19999.000000 19999.000000 19999.000000 19999.000000 19999.000000 \n", "mean 6.453823 7.928996 3.046252 8.338867 3.691935 \n", "std 2.631016 2.080671 2.332500 1.546759 2.567004 \n", "min 0.000000 0.000000 0.000000 0.000000 0.000000 \n", "25% 5.000000 7.000000 1.000000 8.000000 2.000000 \n", "50% 6.000000 8.000000 3.000000 8.000000 3.000000 \n", "75% 8.000000 9.000000 4.000000 9.000000 5.000000 \n", "max 15.000000 15.000000 15.000000 15.000000 15.000000 \n", "\n", " x16 \n", "count 19999.00000 \n", "mean 7.80119 \n", "std 1.61751 \n", "min 0.00000 \n", "25% 7.00000 \n", "50% 8.00000 \n", "75% 9.00000 \n", "max 15.00000 " ] }, "execution_count": 3, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.describe()" ] }, { "cell_type": "code", "execution_count": 4, "id": "bd1b4855-41f2-4d5e-8807-ec9fa1d0bc08", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "\n", "RangeIndex: 19999 entries, 0 to 19998\n", "Data columns (total 17 columns):\n", " # Column Non-Null Count Dtype \n", "--- ------ -------------- ----- \n", " 0 letter 19999 non-null object\n", " 1 x1 19999 non-null int64 \n", " 2 x2 19999 non-null int64 \n", " 3 x3 19999 non-null int64 \n", " 4 x4 19999 non-null int64 \n", " 5 x5 19999 non-null int64 \n", " 6 x6 19999 non-null int64 \n", " 7 x7 19999 non-null int64 \n", " 8 x8 19999 non-null int64 \n", " 9 x9 19999 non-null int64 \n", " 10 x10 19999 non-null int64 \n", " 11 x11 19999 non-null int64 \n", " 12 x12 19999 non-null int64 \n", " 13 x13 19999 non-null int64 \n", " 14 x14 19999 non-null int64 \n", " 15 x15 19999 non-null int64 \n", " 16 x16 19999 non-null int64 \n", "dtypes: int64(16), object(1)\n", "memory usage: 2.6+ MB\n" ] } ], "source": [ "df.info()" ] }, { "cell_type": "code", "execution_count": 5, "id": "0d862011-ea3b-4ed8-97e8-64b8b56b926a", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "letter 0\n", "x1 0\n", "x2 0\n", "x3 0\n", "x4 0\n", "x5 0\n", "x6 0\n", "x7 0\n", "x8 0\n", "x9 0\n", "x10 0\n", "x11 0\n", "x12 0\n", "x13 0\n", "x14 0\n", "x15 0\n", "x16 0\n", "dtype: int64" ] }, "execution_count": 5, "metadata": {}, "output_type": "execute_result" } ], "source": [ "df.isna().sum()" ] }, { "cell_type": "code", "execution_count": 6, "id": "501d63db-86c6-42ba-a6d6-b52ddf629f15", "metadata": {}, "outputs": [], "source": [ "import numpy as npy\n", "import seaborn as sns\n", "import matplotlib.pyplot as plt" ] }, { "cell_type": "code", "execution_count": 7, "id": "11a9ad1f-1792-4d6c-aa50-5ce36605cde0", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axes = plt.subplots(4, 4, figsize=(20, 20))\n", "\n", "for ax, col in zip(axes.flatten(), cols_useful):\n", " sns.boxplot(y = df[col], ax = ax)\n", "\n", "\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 8, "id": "5889e6b1-465a-4f97-8b3f-87e95c012d7c", "metadata": {}, "outputs": [], "source": [ "def iqr(df_column):\n", " q1 = df_column.quantile(0.25)\n", " q3 = df_column.quantile(0.75)\n", " iqr = (q3 - q1) * 1.5\n", " lower = q1 - iqr\n", " upper = q3 + iqr\n", " df_column = df_column.apply(lambda x: q1 if x < lower else (q3 if x > upper else x))\n", " return df_column" ] }, { "cell_type": "code", "execution_count": 9, "id": "3e5259f6-1ad7-45f3-a17b-9d074ebcf698", "metadata": {}, "outputs": [], "source": [ "for c in cols_useful:\n", " df[c] = iqr(df[c])" ] }, { "cell_type": "code", "execution_count": 10, "id": "36290bf6-178f-4ee0-8441-1504da1e3431", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "fig, axes = plt.subplots(4, 4, figsize=(20, 20))\n", "\n", "for ax, col in zip(axes.flatten(), cols_useful):\n", " sns.boxplot(y = df[col], ax = ax)\n", "\n", "\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 12, "id": "b6da4f23-bf61-4155-a7fa-bdaa1ecff42c", "metadata": {}, "outputs": [], "source": [ "from tensorflow import keras\n", "from sklearn.preprocessing import StandardScaler, LabelEncoder\n", "from sklearn.model_selection import train_test_split\n", "\n", "\n", "X = df.drop('letter', axis=1)\n", "Y = df['letter']\n", "\n", "lab = LabelEncoder()\n", "\n", "# std = StandardScaler()\n", "\n", "Y = lab.fit_transform(Y)\n", "# print(Y)\n", "\n", "X = X.values\n", "\n", "# Y = Y.reshape(-1,1) # Dont do this. this is for standard scaler. \n", "\n", "# X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size = 0.2, random_state=42)\n", "\n", "X_train, X_test, y_train, y_test = train_test_split(X, Y, test_size = 0.2, random_state=42)\n", "\n", "# X_train = std.fit_transform(X_train)\n", "# X_test = std.transform(X_test)\n", "\n", "scaler_X = StandardScaler()\n", "X_train = scaler_X.fit_transform(X_train)\n", "X_test = scaler_X.transform(X_test) # we want to normalize based on Alpha and mean of test data for test\"\n" ] }, { "cell_type": "code", "execution_count": 13, "id": "b3299b62-757d-45fd-bf35-71f8079d1276", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(19999,)" ] }, "execution_count": 13, "metadata": {}, "output_type": "execute_result" } ], "source": [ "Y.shape" ] }, { "cell_type": "code", "execution_count": 14, "id": "7ba2737b-172f-4ec6-bf86-5d4dd726fd88", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "(19999, 16)" ] }, "execution_count": 14, "metadata": {}, "output_type": "execute_result" } ], "source": [ "X.shape" ] }, { "cell_type": "code", "execution_count": 15, "id": "01f6958a-1d69-4253-8fdc-d4d5f143cc5c", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/home/soham/anaconda3/envs/jupyter_env/lib/python3.10/site-packages/keras/src/layers/core/dense.py:95: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n", " super().__init__(activity_regularizer=activity_regularizer, **kwargs)\n" ] } ], "source": [ "model_tf = keras.Sequential([\n", " keras.layers.Dense(64, activation='relu', input_shape=(X.shape[1],)),\n", " # keras.layers.Dense(64, activation='relu', input_shape=(X_train.shape[1],)),\n", " keras.layers.Dense(64, activation = 'relu'),\n", " keras.layers.Dense(32, activation = 'relu'),\n", " keras.layers.Dense(26, activation = 'softmax'),\n", "])" ] }, { "cell_type": "code", "execution_count": 16, "id": "c15e7f7f-0daa-4e46-92dd-2a7ec5644987", "metadata": {}, "outputs": [], "source": [ "model_tf.compile(\n", " loss= 'sparse_categorical_crossentropy',\n", " optimizer = 'adam',\n", " metrics = ['accuracy']\n", ")" ] }, { "cell_type": "code", "execution_count": 17, "id": "715e6cb4-c41b-47bb-a8f1-5d1b2800cdd2", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
Model: \"sequential\"\n",
       "
\n" ], "text/plain": [ "\u001b[1mModel: \"sequential\"\u001b[0m\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
       "┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃\n",
       "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
       "│ dense (Dense)                   │ (None, 64)             │         1,088 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_1 (Dense)                 │ (None, 64)             │         4,160 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_2 (Dense)                 │ (None, 32)             │         2,080 │\n",
       "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
       "│ dense_3 (Dense)                 │ (None, 26)             │           858 │\n",
       "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
       "
\n" ], "text/plain": [ "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n", "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n", "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n", "│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m1,088\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ dense_1 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m) │ \u001b[38;5;34m4,160\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ dense_2 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m) │ \u001b[38;5;34m2,080\u001b[0m │\n", "├─────────────────────────────────┼────────────────────────┼───────────────┤\n", "│ dense_3 (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m26\u001b[0m) │ \u001b[38;5;34m858\u001b[0m │\n", "└─────────────────────────────────┴────────────────────────┴───────────────┘\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Total params: 8,186 (31.98 KB)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m8,186\u001b[0m (31.98 KB)\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Trainable params: 8,186 (31.98 KB)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m8,186\u001b[0m (31.98 KB)\n" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "text/html": [ "
 Non-trainable params: 0 (0.00 B)\n",
       "
\n" ], "text/plain": [ "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "model_tf.summary()" ] }, { "cell_type": "code", "execution_count": 18, "id": "a35f3c9a-316f-4997-87d2-db60ee99463a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/10\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "2026-05-13 17:17:12.499016: I external/local_xla/xla/service/service.cc:163] XLA service 0x70f8e8013e90 initialized for platform Host (this does not guarantee that XLA will be used). Devices:\n", "2026-05-13 17:17:12.499027: I external/local_xla/xla/service/service.cc:171] StreamExecutor device (0): Host, Default Version\n", "2026-05-13 17:17:12.513140: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\u001b[1m200/500\u001b[0m \u001b[32m━━━━━━━━\u001b[0m\u001b[37m━━━━━━━━━━━━\u001b[0m \u001b[1m0s\u001b[0m 755us/step - accuracy: 0.1693 - loss: 2.9050 " ] }, { "name": "stderr", "output_type": "stream", "text": [ "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n", "I0000 00:00:1778672832.712514 93460 device_compiler.h:196] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process.\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 2ms/step - accuracy: 0.5113 - loss: 1.7113 - val_accuracy: 0.7045 - val_loss: 1.0469\n", "Epoch 2/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step - accuracy: 0.7460 - loss: 0.8621 - val_accuracy: 0.7688 - val_loss: 0.7898\n", "Epoch 3/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step - accuracy: 0.7936 - loss: 0.6870 - val_accuracy: 0.8040 - val_loss: 0.6552\n", "Epoch 4/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 965us/step - accuracy: 0.8245 - loss: 0.5793 - val_accuracy: 0.8250 - val_loss: 0.5834\n", "Epoch 5/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step - accuracy: 0.8451 - loss: 0.5073 - val_accuracy: 0.8385 - val_loss: 0.5267\n", "Epoch 6/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 994us/step - accuracy: 0.8616 - loss: 0.4563 - val_accuracy: 0.8575 - val_loss: 0.4765\n", "Epoch 7/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 989us/step - accuracy: 0.8734 - loss: 0.4118 - val_accuracy: 0.8635 - val_loss: 0.4466\n", "Epoch 8/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1ms/step - accuracy: 0.8831 - loss: 0.3778 - val_accuracy: 0.8683 - val_loss: 0.4369\n", "Epoch 9/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 975us/step - accuracy: 0.8919 - loss: 0.3509 - val_accuracy: 0.8808 - val_loss: 0.3990\n", "Epoch 10/10\n", "\u001b[1m500/500\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 983us/step - accuracy: 0.8982 - loss: 0.3265 - val_accuracy: 0.8907 - val_loss: 0.3701\n" ] } ], "source": [ "# history = model_tf.fit(\n", "# X_train, y_train,\n", "# validation_data= (X_test, y_test),\n", "# batch_size = 32,\n", "# epochs = 100\n", "# )\n", "# # help(model_tf.fit)\n", "\n", "# Train model\n", "history = model_tf.fit(X_train, y_train, epochs=10, batch_size=32, validation_split=0.1, validation_data= (X_test, y_test))" ] }, { "cell_type": "code", "execution_count": 22, "id": "8ea9f09b-0953-40f9-b11e-2cafc73d3661", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[]" ] }, "execution_count": 22, "metadata": {}, "output_type": "execute_result" }, { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# help(plt.plot)\n", "\n", "plt.plot(history.history[\"accuracy\"], label = \"accuracy\")\n", "plt.plot(history.history[\"val_accuracy\"], label= \"Val accuracy\")" ] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "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": "3.10.16" } }, "nbformat": 4, "nbformat_minor": 5 }