diff --git a/src/classes/Changelog.py b/src/classes/Changelog.py index 4cbec067..6b04226e 100644 --- a/src/classes/Changelog.py +++ b/src/classes/Changelog.py @@ -7,7 +7,7 @@ from classes.ColorText import colorText -VERSION = "2.15" +VERSION = "2.16" changelog = colorText.BOLD + '[ChangeLog]\n' + colorText.END + colorText.BLUE + ''' [1.00 - Beta] @@ -271,4 +271,7 @@ [2.15] 1. MA Reversal improved for trend following (Inspired from Siddhart Bhanushali's 44 SMA) +[2.16] +1. Nifty Prediction NaN values handled gracefully with forward filling if data is absent +2. Ticker 0 > Search by Stock name - re-enabled in GUI ''' + colorText.END diff --git a/src/classes/Screener.py b/src/classes/Screener.py index 55f735d1..6ec12597 100644 --- a/src/classes/Screener.py +++ b/src/classes/Screener.py @@ -621,6 +621,7 @@ def getNiftyPrediction(self, data, proxyServer): ### v2 Preprocessing for col in pkl['columns']: data[col] = data[col].pct_change(fill_method=None) * 100 + data = data.ffill().dropna() data = data.iloc[-1] ### data = pkl['scaler'].transform([data]) diff --git a/src/ml/experiment.ipynb b/src/ml/experiment.ipynb index 10e48c77..4604af1f 100644 --- a/src/ml/experiment.ipynb +++ b/src/ml/experiment.ipynb @@ -2,7 +2,7 @@ "cells": [ { "cell_type": "code", - "execution_count": 1, + "execution_count": 11, "metadata": {}, "outputs": [], "source": [ @@ -21,7 +21,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 12, "metadata": {}, "outputs": [], "source": [ @@ -37,17 +37,17 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 13, "metadata": {}, "outputs": [], "source": [ - "TEST_DAYS = 50\n", + "TEST_DAYS = 10\n", "PERIOD = '5y'" ] }, { "cell_type": "code", - "execution_count": 4, + "execution_count": 14, "metadata": {}, "outputs": [], "source": [ @@ -56,7 +56,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 15, "metadata": {}, "outputs": [], "source": [ @@ -65,7 +65,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -126,7 +126,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": 23, "metadata": {}, "outputs": [], "source": [ @@ -146,12 +146,14 @@ " df['crude_Low'] = df['crude_Low'].pct_change(fill_method=None) * 100\n", " df['crude_Open'] = df['crude_Open'].pct_change(fill_method=None) * 100\n", " df['crude_Close'] = df['crude_Close'].pct_change(fill_method=None) * 100\n", + " \n", + " df = df.ffill().dropna()\n", " return df" ] }, { "cell_type": "code", - "execution_count": 8, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -160,7 +162,7 @@ }, { "cell_type": "code", - "execution_count": 9, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -169,7 +171,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "metadata": {}, "outputs": [], "source": [ @@ -187,225 +189,27 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": null, "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", - "
OpenHighLowClosegold_Closecrude_Close
Date
2018-11-190.8197110.7437930.5422690.7601450.1883950.531348
2018-11-200.082466-0.314167-0.448602-0.995970-0.335212-5.866804
2018-11-21-0.643843-0.647526-0.737723-0.5269270.5578382.245931
2018-11-270.5029230.5391211.0162260.536289-0.737584-0.135579
2018-11-280.8219200.5857740.9767510.4047500.842146-2.463151
.....................
2023-08-070.5859840.4176280.4545660.411440-0.314497-1.062542
2023-08-080.2571900.0711050.042504-0.134974-0.4861661.195992
2023-08-09-0.2465890.056531-0.3358380.315271-0.4521571.784857
2023-08-100.136627-0.1114780.143318-0.455627-0.052208-1.872040
2023-08-11-0.261665-0.335563-0.423948-0.587414-0.0783540.446755
\n", - "

966 rows × 6 columns

\n", - "
" - ], - "text/plain": [ - " Open High Low Close gold_Close crude_Close\n", - "Date \n", - "2018-11-19 0.819711 0.743793 0.542269 0.760145 0.188395 0.531348\n", - "2018-11-20 0.082466 -0.314167 -0.448602 -0.995970 -0.335212 -5.866804\n", - "2018-11-21 -0.643843 -0.647526 -0.737723 -0.526927 0.557838 2.245931\n", - "2018-11-27 0.502923 0.539121 1.016226 0.536289 -0.737584 -0.135579\n", - "2018-11-28 0.821920 0.585774 0.976751 0.404750 0.842146 -2.463151\n", - "... ... ... ... ... ... ...\n", - "2023-08-07 0.585984 0.417628 0.454566 0.411440 -0.314497 -1.062542\n", - "2023-08-08 0.257190 0.071105 0.042504 -0.134974 -0.486166 1.195992\n", - "2023-08-09 -0.246589 0.056531 -0.335838 0.315271 -0.452157 1.784857\n", - "2023-08-10 0.136627 -0.111478 0.143318 -0.455627 -0.052208 -1.872040\n", - "2023-08-11 -0.261665 -0.335563 -0.423948 -0.587414 -0.078354 0.446755\n", - "\n", - "[966 rows x 6 columns]" - ] - }, - "execution_count": 11, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "x" ] }, { "cell_type": "code", - "execution_count": 12, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "Date\n", - "2018-11-19 1.0\n", - "2018-11-20 1.0\n", - "2018-11-21 1.0\n", - "2018-11-27 0.0\n", - "2018-11-28 0.0\n", - " ... \n", - "2023-08-07 1.0\n", - "2023-08-08 0.0\n", - "2023-08-09 1.0\n", - "2023-08-10 1.0\n", - "2023-08-11 1.0\n", - "Name: target, Length: 966, dtype: float64" - ] - }, - "execution_count": 12, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "y" ] }, { "cell_type": "code", - "execution_count": 13, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "No. of Bullish samples: 506\n", - "No. of Bearish samples: 460\n" - ] - } - ], + "outputs": [], "source": [ "print('No. of Bullish samples: {}'.format(y[y == 0].size))\n", "print('No. of Bearish samples: {}'.format(y[y == 1].size))" @@ -413,39 +217,9 @@ }, { "cell_type": "code", - "execution_count": 14, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Using StandardScaler\n" - ] - }, - { - "data": { - "text/plain": [ - "array([[ 6.58597736e-01, 8.04494879e-01, 4.51528068e-01,\n", - " 7.66463042e-01, 1.35910062e-01, 5.93388622e-02],\n", - " [ 1.19662220e-02, -4.34193703e-01, -5.45172893e-01,\n", - " -1.19559615e+00, -4.18790357e-01, -5.59228023e-01],\n", - " [-6.25073637e-01, -8.24499116e-01, -8.35994774e-01,\n", - " -6.71547487e-01, 5.27292121e-01, 2.25102944e-01],\n", - " ...,\n", - " [-2.76644732e-01, -1.69660885e-04, -4.31745390e-01,\n", - " 2.69417275e-01, -5.42680065e-01, 1.80526806e-01],\n", - " [ 5.94709103e-02, -1.96879502e-01, 5.02297622e-02,\n", - " -5.91885808e-01, -1.18981251e-01, -1.73018278e-01],\n", - " [-2.89867741e-01, -4.59244766e-01, -5.20373755e-01,\n", - " -7.39127604e-01, -1.46678952e-01, 5.11604994e-02]])" - ] - }, - "execution_count": 14, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "if not INDICATOR_DATASET:\n", " print(\"Using StandardScaler\")\n", @@ -465,59 +239,18 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "[PhysicalDevice(name='/physical_device:CPU:0', device_type='CPU')]" - ] - }, - "execution_count": 15, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "visible_devices" ] }, { "cell_type": "code", - "execution_count": 119, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Model: \"sequential_20\"\n", - "_________________________________________________________________\n", - " Layer (type) Output Shape Param # \n", - "=================================================================\n", - " dense_140 (Dense) (None, 64) 448 \n", - " \n", - " dense_141 (Dense) (None, 32) 2080 \n", - " \n", - " dense_142 (Dense) (None, 16) 528 \n", - " \n", - " dense_143 (Dense) (None, 8) 136 \n", - " \n", - " dense_144 (Dense) (None, 4) 36 \n", - " \n", - " dense_145 (Dense) (None, 2) 10 \n", - " \n", - " dense_146 (Dense) (None, 1) 3 \n", - " \n", - "=================================================================\n", - "Total params: 3241 (12.66 KB)\n", - "Trainable params: 3241 (12.66 KB)\n", - "Non-trainable params: 0 (0.00 Byte)\n", - "_________________________________________________________________\n" - ] - } - ], + "outputs": [], "source": [ "import tensorflow as tf\n", "from keras import Sequential\n", @@ -572,1295 +305,9 @@ }, { "cell_type": "code", - "execution_count": 120, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "BATCH SIZE = 256\n", - "Epoch 1/750\n", - "\n", - "Epoch 1: val_accuracy improved from -inf to 0.46207, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.6894 - accuracy: 0.5347 - val_loss: 0.7032 - val_accuracy: 0.4621 - lr: 0.0010 - 258ms/epoch - 64ms/step\n", - "Epoch 2/750\n", - "\n", - "Epoch 2: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6886 - accuracy: 0.5347 - val_loss: 0.7032 - val_accuracy: 0.4621 - lr: 0.0010 - 15ms/epoch - 4ms/step\n", - "Epoch 3/750\n", - "\n", - "Epoch 3: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6874 - accuracy: 0.5347 - val_loss: 0.7030 - val_accuracy: 0.4621 - lr: 9.9750e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 4/750\n", - "\n", - "Epoch 4: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6854 - accuracy: 0.5347 - val_loss: 0.7024 - val_accuracy: 0.4621 - lr: 9.9501e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 5/750\n", - "\n", - "Epoch 5: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6826 - accuracy: 0.5347 - val_loss: 0.7017 - val_accuracy: 0.4621 - lr: 9.9253e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 6/750\n", - "\n", - "Epoch 6: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6795 - accuracy: 0.5347 - val_loss: 0.7012 - val_accuracy: 0.4621 - lr: 9.9005e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 7/750\n", - "\n", - "Epoch 7: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6761 - accuracy: 0.5347 - val_loss: 0.7005 - val_accuracy: 0.4621 - lr: 9.8758e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 8/750\n", - "\n", - "Epoch 8: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6727 - accuracy: 0.5347 - val_loss: 0.7000 - val_accuracy: 0.4621 - lr: 9.8511e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 9/750\n", - "\n", - "Epoch 9: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6697 - accuracy: 0.5347 - val_loss: 0.6997 - val_accuracy: 0.4621 - lr: 9.8265e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 10/750\n", - "\n", - "Epoch 10: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6664 - accuracy: 0.5347 - val_loss: 0.6994 - val_accuracy: 0.4621 - lr: 9.8020e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 11/750\n", - "\n", - "Epoch 11: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6636 - accuracy: 0.5347 - val_loss: 0.6989 - val_accuracy: 0.4621 - lr: 9.7775e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 12/750\n", - "\n", - "Epoch 12: val_accuracy did not improve from 0.46207\n", - "4/4 - 0s - loss: 0.6608 - accuracy: 0.5347 - val_loss: 0.6984 - val_accuracy: 0.4621 - lr: 9.7531e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 13/750\n", - "\n", - "Epoch 13: val_accuracy improved from 0.46207 to 0.60690, saving model to best_model.h5\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/Users/pranjaljoshi/miniforge3/envs/screenipy/lib/python3.10/site-packages/keras/src/engine/training.py:3079: UserWarning: You are saving your model as an HDF5 file via `model.save()`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')`.\n", - " saving_api.save_model(\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "4/4 - 0s - loss: 0.6584 - accuracy: 0.5347 - val_loss: 0.6979 - val_accuracy: 0.6069 - lr: 9.7287e-04 - 38ms/epoch - 10ms/step\n", - "Epoch 14/750\n", - "\n", - "Epoch 14: val_accuracy did not improve from 0.60690\n", - "4/4 - 0s - loss: 0.6563 - accuracy: 0.6590 - val_loss: 0.6973 - val_accuracy: 0.6069 - lr: 9.7045e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 15/750\n", - "\n", - "Epoch 15: val_accuracy improved from 0.60690 to 0.61379, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.6544 - accuracy: 0.6565 - val_loss: 0.6967 - val_accuracy: 0.6138 - lr: 9.6802e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 16/750\n", - "\n", - "Epoch 16: val_accuracy did not improve from 0.61379\n", - "4/4 - 0s - loss: 0.6526 - accuracy: 0.6577 - val_loss: 0.6962 - val_accuracy: 0.6138 - lr: 9.6561e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 17/750\n", - "\n", - "Epoch 17: val_accuracy did not improve from 0.61379\n", - "4/4 - 0s - loss: 0.6510 - accuracy: 0.6590 - val_loss: 0.6956 - val_accuracy: 0.6138 - lr: 9.6319e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 18/750\n", - "\n", - "Epoch 18: val_accuracy improved from 0.61379 to 0.62069, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.6493 - accuracy: 0.6614 - val_loss: 0.6949 - val_accuracy: 0.6207 - lr: 9.6079e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 19/750\n", - "\n", - "Epoch 19: val_accuracy did not improve from 0.62069\n", - "4/4 - 0s - loss: 0.6479 - accuracy: 0.6614 - val_loss: 0.6942 - val_accuracy: 0.6207 - lr: 9.5839e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 20/750\n", - "\n", - "Epoch 20: val_accuracy improved from 0.62069 to 0.62759, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.6467 - accuracy: 0.6577 - val_loss: 0.6935 - val_accuracy: 0.6276 - lr: 9.5600e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 21/750\n", - "\n", - "Epoch 21: val_accuracy did not improve from 0.62759\n", - "4/4 - 0s - loss: 0.6454 - accuracy: 0.6541 - val_loss: 0.6928 - val_accuracy: 0.6276 - lr: 9.5361e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 22/750\n", - "\n", - "Epoch 22: val_accuracy did not improve from 0.62759\n", - "4/4 - 0s - loss: 0.6439 - accuracy: 0.6577 - val_loss: 0.6920 - val_accuracy: 0.6276 - lr: 9.5123e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 23/750\n", - "\n", - "Epoch 23: val_accuracy did not improve from 0.62759\n", - "4/4 - 0s - loss: 0.6427 - accuracy: 0.6577 - val_loss: 0.6914 - val_accuracy: 0.6276 - lr: 9.4885e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 24/750\n", - "\n", - "Epoch 24: val_accuracy improved from 0.62759 to 0.63448, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.6415 - accuracy: 0.6614 - val_loss: 0.6907 - val_accuracy: 0.6345 - lr: 9.4648e-04 - 22ms/epoch - 5ms/step\n", - "Epoch 25/750\n", - "\n", - "Epoch 25: val_accuracy did not improve from 0.63448\n", - "4/4 - 0s - loss: 0.6404 - accuracy: 0.6650 - val_loss: 0.6900 - val_accuracy: 0.6345 - lr: 9.4412e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 26/750\n", - "\n", - "Epoch 26: val_accuracy did not improve from 0.63448\n", - "4/4 - 0s - loss: 0.6392 - accuracy: 0.6699 - val_loss: 0.6894 - val_accuracy: 0.6276 - lr: 9.4176e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 27/750\n", - "\n", - "Epoch 27: val_accuracy did not improve from 0.63448\n", - "4/4 - 0s - loss: 0.6380 - accuracy: 0.6724 - val_loss: 0.6888 - val_accuracy: 0.6276 - lr: 9.3941e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 28/750\n", - "\n", - "Epoch 28: val_accuracy did not improve from 0.63448\n", - "4/4 - 0s - loss: 0.6369 - accuracy: 0.6736 - val_loss: 0.6882 - val_accuracy: 0.6276 - lr: 9.3707e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 29/750\n", - "\n", - "Epoch 29: val_accuracy did not improve from 0.63448\n", - "4/4 - 0s - loss: 0.6358 - accuracy: 0.6748 - val_loss: 0.6875 - val_accuracy: 0.6276 - lr: 9.3473e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 30/750\n", - "\n", - "Epoch 30: val_accuracy did not improve from 0.63448\n", - "4/4 - 0s - loss: 0.6346 - accuracy: 0.6784 - val_loss: 0.6867 - val_accuracy: 0.6345 - lr: 9.3239e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 31/750\n", - "\n", - "Epoch 31: val_accuracy improved from 0.63448 to 0.64138, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.6336 - accuracy: 0.6772 - val_loss: 0.6859 - val_accuracy: 0.6414 - lr: 9.3007e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 32/750\n", - "\n", - "Epoch 32: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6325 - accuracy: 0.6772 - val_loss: 0.6851 - val_accuracy: 0.6414 - lr: 9.2774e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 33/750\n", - "\n", - "Epoch 33: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6315 - accuracy: 0.6821 - val_loss: 0.6843 - val_accuracy: 0.6345 - lr: 9.2543e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 34/750\n", - "\n", - "Epoch 34: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6306 - accuracy: 0.6882 - val_loss: 0.6836 - val_accuracy: 0.6276 - lr: 9.2312e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 35/750\n", - "\n", - "Epoch 35: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6298 - accuracy: 0.6955 - val_loss: 0.6829 - val_accuracy: 0.6345 - lr: 9.2081e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 36/750\n", - "\n", - "Epoch 36: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6289 - accuracy: 0.6955 - val_loss: 0.6822 - val_accuracy: 0.6345 - lr: 9.1851e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 37/750\n", - "\n", - "Epoch 37: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6282 - accuracy: 0.6991 - val_loss: 0.6816 - val_accuracy: 0.6345 - lr: 9.1622e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 38/750\n", - "\n", - "Epoch 38: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6274 - accuracy: 0.6991 - val_loss: 0.6811 - val_accuracy: 0.6276 - lr: 9.1393e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 39/750\n", - "\n", - "Epoch 39: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6266 - accuracy: 0.6991 - val_loss: 0.6804 - val_accuracy: 0.6207 - lr: 9.1165e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 40/750\n", - "\n", - "Epoch 40: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6257 - accuracy: 0.6955 - val_loss: 0.6797 - val_accuracy: 0.6207 - lr: 9.0937e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 41/750\n", - "\n", - "Epoch 41: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6250 - accuracy: 0.6943 - val_loss: 0.6791 - val_accuracy: 0.6276 - lr: 9.0710e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 42/750\n", - "\n", - "Epoch 42: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6242 - accuracy: 0.6967 - val_loss: 0.6784 - val_accuracy: 0.6276 - lr: 9.0484e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 43/750\n", - "\n", - "Epoch 43: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6235 - accuracy: 0.6991 - val_loss: 0.6777 - val_accuracy: 0.6276 - lr: 9.0258e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 44/750\n", - "\n", - "Epoch 44: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6227 - accuracy: 0.7004 - val_loss: 0.6771 - val_accuracy: 0.6276 - lr: 9.0032e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 45/750\n", - "\n", - "Epoch 45: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6218 - accuracy: 0.7028 - val_loss: 0.6766 - val_accuracy: 0.6345 - lr: 8.9808e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 46/750\n", - "\n", - "Epoch 46: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6211 - accuracy: 0.7040 - val_loss: 0.6760 - val_accuracy: 0.6414 - lr: 8.9583e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 47/750\n", - "\n", - "Epoch 47: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6204 - accuracy: 0.7077 - val_loss: 0.6755 - val_accuracy: 0.6414 - lr: 8.9360e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 48/750\n", - "\n", - "Epoch 48: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6197 - accuracy: 0.7089 - val_loss: 0.6751 - val_accuracy: 0.6414 - lr: 8.9137e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 49/750\n", - "\n", - "Epoch 49: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6190 - accuracy: 0.7077 - val_loss: 0.6747 - val_accuracy: 0.6414 - lr: 8.8914e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 50/750\n", - "\n", - "Epoch 50: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6184 - accuracy: 0.7101 - val_loss: 0.6743 - val_accuracy: 0.6345 - lr: 8.8692e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 51/750\n", - "\n", - "Epoch 51: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6177 - accuracy: 0.7101 - val_loss: 0.6739 - val_accuracy: 0.6345 - lr: 8.8471e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 52/750\n", - "\n", - "Epoch 52: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6172 - accuracy: 0.7125 - val_loss: 0.6736 - val_accuracy: 0.6345 - lr: 8.8250e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 53/750\n", - "\n", - "Epoch 53: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6166 - accuracy: 0.7125 - val_loss: 0.6732 - val_accuracy: 0.6345 - lr: 8.8029e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 54/750\n", - "\n", - "Epoch 54: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6161 - accuracy: 0.7125 - val_loss: 0.6728 - val_accuracy: 0.6345 - lr: 8.7809e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 55/750\n", - "\n", - "Epoch 55: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6155 - accuracy: 0.7113 - val_loss: 0.6723 - val_accuracy: 0.6345 - lr: 8.7590e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 56/750\n", - "\n", - "Epoch 56: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6150 - accuracy: 0.7101 - val_loss: 0.6719 - val_accuracy: 0.6414 - lr: 8.7372e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 57/750\n", - "\n", - "Epoch 57: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6144 - accuracy: 0.7113 - val_loss: 0.6716 - val_accuracy: 0.6345 - lr: 8.7153e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 58/750\n", - "\n", - "Epoch 58: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6138 - accuracy: 0.7125 - val_loss: 0.6712 - val_accuracy: 0.6345 - lr: 8.6936e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 59/750\n", - "\n", - "Epoch 59: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6132 - accuracy: 0.7138 - val_loss: 0.6708 - val_accuracy: 0.6345 - lr: 8.6719e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 60/750\n", - "\n", - "Epoch 60: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6124 - accuracy: 0.7162 - val_loss: 0.6704 - val_accuracy: 0.6345 - lr: 8.6502e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 61/750\n", - "\n", - "Epoch 61: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6119 - accuracy: 0.7138 - val_loss: 0.6699 - val_accuracy: 0.6345 - lr: 8.6286e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 62/750\n", - "\n", - "Epoch 62: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6112 - accuracy: 0.7174 - val_loss: 0.6695 - val_accuracy: 0.6345 - lr: 8.6071e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 63/750\n", - "\n", - "Epoch 63: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6107 - accuracy: 0.7186 - val_loss: 0.6691 - val_accuracy: 0.6345 - lr: 8.5856e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 64/750\n", - "\n", - "Epoch 64: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6101 - accuracy: 0.7199 - val_loss: 0.6687 - val_accuracy: 0.6345 - lr: 8.5641e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 65/750\n", - "\n", - "Epoch 65: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6096 - accuracy: 0.7199 - val_loss: 0.6684 - val_accuracy: 0.6345 - lr: 8.5428e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 66/750\n", - "\n", - "Epoch 66: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6090 - accuracy: 0.7174 - val_loss: 0.6681 - val_accuracy: 0.6345 - lr: 8.5214e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 67/750\n", - "\n", - "Epoch 67: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6087 - accuracy: 0.7162 - val_loss: 0.6679 - val_accuracy: 0.6345 - lr: 8.5002e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 68/750\n", - "\n", - "Epoch 68: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6081 - accuracy: 0.7174 - val_loss: 0.6675 - val_accuracy: 0.6276 - lr: 8.4789e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 69/750\n", - "\n", - "Epoch 69: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6075 - accuracy: 0.7199 - val_loss: 0.6673 - val_accuracy: 0.6345 - lr: 8.4578e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 70/750\n", - "\n", - "Epoch 70: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6070 - accuracy: 0.7186 - val_loss: 0.6671 - val_accuracy: 0.6345 - lr: 8.4366e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 71/750\n", - "\n", - "Epoch 71: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6065 - accuracy: 0.7211 - val_loss: 0.6668 - val_accuracy: 0.6414 - lr: 8.4156e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 72/750\n", - "\n", - "Epoch 72: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6061 - accuracy: 0.7199 - val_loss: 0.6664 - val_accuracy: 0.6414 - lr: 8.3946e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 73/750\n", - "\n", - "Epoch 73: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6056 - accuracy: 0.7211 - val_loss: 0.6662 - val_accuracy: 0.6414 - lr: 8.3736e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 74/750\n", - "\n", - "Epoch 74: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6051 - accuracy: 0.7211 - val_loss: 0.6659 - val_accuracy: 0.6414 - lr: 8.3527e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 75/750\n", - "\n", - "Epoch 75: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6046 - accuracy: 0.7235 - val_loss: 0.6656 - val_accuracy: 0.6345 - lr: 8.3318e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 76/750\n", - "\n", - "Epoch 76: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6042 - accuracy: 0.7235 - val_loss: 0.6653 - val_accuracy: 0.6345 - lr: 8.3110e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 77/750\n", - "\n", - "Epoch 77: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6038 - accuracy: 0.7235 - val_loss: 0.6651 - val_accuracy: 0.6345 - lr: 8.2903e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 78/750\n", - "\n", - "Epoch 78: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6033 - accuracy: 0.7247 - val_loss: 0.6648 - val_accuracy: 0.6276 - lr: 8.2696e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 79/750\n", - "\n", - "Epoch 79: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6028 - accuracy: 0.7259 - val_loss: 0.6645 - val_accuracy: 0.6276 - lr: 8.2489e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 80/750\n", - "\n", - "Epoch 80: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6025 - accuracy: 0.7272 - val_loss: 0.6642 - val_accuracy: 0.6345 - lr: 8.2283e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 81/750\n", - "\n", - "Epoch 81: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6021 - accuracy: 0.7272 - val_loss: 0.6640 - val_accuracy: 0.6345 - lr: 8.2078e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 82/750\n", - "\n", - "Epoch 82: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6017 - accuracy: 0.7284 - val_loss: 0.6639 - val_accuracy: 0.6276 - lr: 8.1873e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 83/750\n", - "\n", - "Epoch 83: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6012 - accuracy: 0.7296 - val_loss: 0.6637 - val_accuracy: 0.6345 - lr: 8.1669e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 84/750\n", - "\n", - "Epoch 84: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6008 - accuracy: 0.7333 - val_loss: 0.6635 - val_accuracy: 0.6345 - lr: 8.1465e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 85/750\n", - "\n", - "Epoch 85: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6005 - accuracy: 0.7333 - val_loss: 0.6633 - val_accuracy: 0.6345 - lr: 8.1261e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 86/750\n", - "\n", - "Epoch 86: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.6001 - accuracy: 0.7333 - val_loss: 0.6631 - val_accuracy: 0.6345 - lr: 8.1058e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 87/750\n", - "\n", - "Epoch 87: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5997 - accuracy: 0.7333 - val_loss: 0.6628 - val_accuracy: 0.6345 - lr: 8.0856e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 88/750\n", - "\n", - "Epoch 88: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5994 - accuracy: 0.7333 - val_loss: 0.6627 - val_accuracy: 0.6345 - lr: 8.0654e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 89/750\n", - "\n", - "Epoch 89: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5991 - accuracy: 0.7333 - val_loss: 0.6625 - val_accuracy: 0.6345 - lr: 8.0453e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 90/750\n", - "\n", - "Epoch 90: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5987 - accuracy: 0.7333 - val_loss: 0.6622 - val_accuracy: 0.6345 - lr: 8.0252e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 91/750\n", - "\n", - "Epoch 91: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5984 - accuracy: 0.7320 - val_loss: 0.6620 - val_accuracy: 0.6345 - lr: 8.0051e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 92/750\n", - "\n", - "Epoch 92: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5981 - accuracy: 0.7296 - val_loss: 0.6618 - val_accuracy: 0.6345 - lr: 7.9851e-04 - 36ms/epoch - 9ms/step\n", - "Epoch 93/750\n", - "\n", - "Epoch 93: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5977 - accuracy: 0.7284 - val_loss: 0.6615 - val_accuracy: 0.6345 - lr: 7.9652e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 94/750\n", - "\n", - "Epoch 94: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5974 - accuracy: 0.7296 - val_loss: 0.6613 - val_accuracy: 0.6414 - lr: 7.9453e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 95/750\n", - "\n", - "Epoch 95: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5971 - accuracy: 0.7296 - val_loss: 0.6610 - val_accuracy: 0.6414 - lr: 7.9255e-04 - 18ms/epoch - 5ms/step\n", - "Epoch 96/750\n", - "\n", - "Epoch 96: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5968 - accuracy: 0.7320 - val_loss: 0.6608 - val_accuracy: 0.6414 - lr: 7.9057e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 97/750\n", - "\n", - "Epoch 97: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5965 - accuracy: 0.7320 - val_loss: 0.6606 - val_accuracy: 0.6414 - lr: 7.8860e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 98/750\n", - "\n", - "Epoch 98: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5961 - accuracy: 0.7320 - val_loss: 0.6605 - val_accuracy: 0.6414 - lr: 7.8663e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 99/750\n", - "\n", - "Epoch 99: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5958 - accuracy: 0.7296 - val_loss: 0.6603 - val_accuracy: 0.6414 - lr: 7.8466e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 100/750\n", - "\n", - "Epoch 100: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5954 - accuracy: 0.7296 - val_loss: 0.6602 - val_accuracy: 0.6414 - lr: 7.8270e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 101/750\n", - "\n", - "Epoch 101: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5951 - accuracy: 0.7296 - val_loss: 0.6600 - val_accuracy: 0.6414 - lr: 7.8075e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 102/750\n", - "\n", - "Epoch 102: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5948 - accuracy: 0.7296 - val_loss: 0.6599 - val_accuracy: 0.6414 - lr: 7.7880e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 103/750\n", - "\n", - "Epoch 103: val_accuracy did not improve from 0.64138\n", - "4/4 - 0s - loss: 0.5946 - accuracy: 0.7308 - val_loss: 0.6596 - val_accuracy: 0.6414 - lr: 7.7685e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 104/750\n", - "\n", - "Epoch 104: val_accuracy improved from 0.64138 to 0.64828, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.5942 - accuracy: 0.7320 - val_loss: 0.6593 - val_accuracy: 0.6483 - lr: 7.7492e-04 - 22ms/epoch - 6ms/step\n", - "Epoch 105/750\n", - "\n", - "Epoch 105: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5940 - accuracy: 0.7320 - val_loss: 0.6591 - val_accuracy: 0.6414 - lr: 7.7298e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 106/750\n", - "\n", - "Epoch 106: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5938 - accuracy: 0.7296 - val_loss: 0.6589 - val_accuracy: 0.6414 - lr: 7.7105e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 107/750\n", - "\n", - "Epoch 107: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5936 - accuracy: 0.7296 - val_loss: 0.6589 - val_accuracy: 0.6414 - lr: 7.6913e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 108/750\n", - "\n", - "Epoch 108: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5933 - accuracy: 0.7308 - val_loss: 0.6587 - val_accuracy: 0.6345 - lr: 7.6720e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 109/750\n", - "\n", - "Epoch 109: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5931 - accuracy: 0.7308 - val_loss: 0.6586 - val_accuracy: 0.6345 - lr: 7.6529e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 110/750\n", - "\n", - "Epoch 110: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5928 - accuracy: 0.7308 - val_loss: 0.6584 - val_accuracy: 0.6345 - lr: 7.6338e-04 - 20ms/epoch - 5ms/step\n", - "Epoch 111/750\n", - "\n", - "Epoch 111: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5925 - accuracy: 0.7308 - val_loss: 0.6583 - val_accuracy: 0.6345 - lr: 7.6147e-04 - 18ms/epoch - 5ms/step\n", - "Epoch 112/750\n", - "\n", - "Epoch 112: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5922 - accuracy: 0.7308 - val_loss: 0.6581 - val_accuracy: 0.6345 - lr: 7.5957e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 113/750\n", - "\n", - "Epoch 113: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5919 - accuracy: 0.7333 - val_loss: 0.6579 - val_accuracy: 0.6345 - lr: 7.5767e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 114/750\n", - "\n", - "Epoch 114: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5917 - accuracy: 0.7308 - val_loss: 0.6577 - val_accuracy: 0.6345 - lr: 7.5578e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 115/750\n", - "\n", - "Epoch 115: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5914 - accuracy: 0.7320 - val_loss: 0.6576 - val_accuracy: 0.6345 - lr: 7.5390e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 116/750\n", - "\n", - "Epoch 116: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5911 - accuracy: 0.7320 - val_loss: 0.6575 - val_accuracy: 0.6345 - lr: 7.5201e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 117/750\n", - "\n", - "Epoch 117: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5908 - accuracy: 0.7345 - val_loss: 0.6575 - val_accuracy: 0.6345 - lr: 7.5014e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 118/750\n", - "\n", - "Epoch 118: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5906 - accuracy: 0.7357 - val_loss: 0.6575 - val_accuracy: 0.6345 - lr: 7.4826e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 119/750\n", - "\n", - "Epoch 119: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5902 - accuracy: 0.7357 - val_loss: 0.6574 - val_accuracy: 0.6345 - lr: 7.4639e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 120/750\n", - "\n", - "Epoch 120: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5900 - accuracy: 0.7369 - val_loss: 0.6573 - val_accuracy: 0.6345 - lr: 7.4453e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 121/750\n", - "\n", - "Epoch 121: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5898 - accuracy: 0.7369 - val_loss: 0.6572 - val_accuracy: 0.6345 - lr: 7.4267e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 122/750\n", - "\n", - "Epoch 122: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5895 - accuracy: 0.7369 - val_loss: 0.6570 - val_accuracy: 0.6414 - lr: 7.4082e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 123/750\n", - "\n", - "Epoch 123: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5892 - accuracy: 0.7357 - val_loss: 0.6568 - val_accuracy: 0.6414 - lr: 7.3897e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 124/750\n", - "\n", - "Epoch 124: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5890 - accuracy: 0.7357 - val_loss: 0.6567 - val_accuracy: 0.6414 - lr: 7.3712e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 125/750\n", - "\n", - "Epoch 125: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5887 - accuracy: 0.7357 - val_loss: 0.6566 - val_accuracy: 0.6414 - lr: 7.3528e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 126/750\n", - "\n", - "Epoch 126: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5885 - accuracy: 0.7357 - val_loss: 0.6564 - val_accuracy: 0.6414 - lr: 7.3345e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 127/750\n", - "\n", - "Epoch 127: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5882 - accuracy: 0.7357 - val_loss: 0.6563 - val_accuracy: 0.6414 - lr: 7.3161e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 128/750\n", - "\n", - "Epoch 128: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5880 - accuracy: 0.7357 - val_loss: 0.6563 - val_accuracy: 0.6414 - lr: 7.2979e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 129/750\n", - "\n", - "Epoch 129: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5877 - accuracy: 0.7357 - val_loss: 0.6562 - val_accuracy: 0.6414 - lr: 7.2797e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 130/750\n", - "\n", - "Epoch 130: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5876 - accuracy: 0.7369 - val_loss: 0.6561 - val_accuracy: 0.6414 - lr: 7.2615e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 131/750\n", - "\n", - "Epoch 131: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5873 - accuracy: 0.7357 - val_loss: 0.6559 - val_accuracy: 0.6414 - lr: 7.2433e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 132/750\n", - "\n", - "Epoch 132: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5871 - accuracy: 0.7357 - val_loss: 0.6558 - val_accuracy: 0.6414 - lr: 7.2253e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 133/750\n", - "\n", - "Epoch 133: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5868 - accuracy: 0.7357 - val_loss: 0.6557 - val_accuracy: 0.6414 - lr: 7.2072e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 134/750\n", - "\n", - "Epoch 134: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5866 - accuracy: 0.7357 - val_loss: 0.6555 - val_accuracy: 0.6414 - lr: 7.1892e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 135/750\n", - "\n", - "Epoch 135: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5864 - accuracy: 0.7357 - val_loss: 0.6554 - val_accuracy: 0.6483 - lr: 7.1713e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 136/750\n", - "\n", - "Epoch 136: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5862 - accuracy: 0.7369 - val_loss: 0.6552 - val_accuracy: 0.6483 - lr: 7.1534e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 137/750\n", - "\n", - "Epoch 137: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5860 - accuracy: 0.7357 - val_loss: 0.6550 - val_accuracy: 0.6483 - lr: 7.1355e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 138/750\n", - "\n", - "Epoch 138: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5858 - accuracy: 0.7345 - val_loss: 0.6549 - val_accuracy: 0.6483 - lr: 7.1177e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 139/750\n", - "\n", - "Epoch 139: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5855 - accuracy: 0.7345 - val_loss: 0.6548 - val_accuracy: 0.6483 - lr: 7.0999e-04 - 31ms/epoch - 8ms/step\n", - "Epoch 140/750\n", - "\n", - "Epoch 140: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5854 - accuracy: 0.7369 - val_loss: 0.6547 - val_accuracy: 0.6483 - lr: 7.0822e-04 - 20ms/epoch - 5ms/step\n", - "Epoch 141/750\n", - "\n", - "Epoch 141: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5851 - accuracy: 0.7369 - val_loss: 0.6546 - val_accuracy: 0.6483 - lr: 7.0645e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 142/750\n", - "\n", - "Epoch 142: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5849 - accuracy: 0.7369 - val_loss: 0.6546 - val_accuracy: 0.6483 - lr: 7.0469e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 143/750\n", - "\n", - "Epoch 143: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5847 - accuracy: 0.7357 - val_loss: 0.6545 - val_accuracy: 0.6483 - lr: 7.0293e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 144/750\n", - "\n", - "Epoch 144: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5845 - accuracy: 0.7345 - val_loss: 0.6543 - val_accuracy: 0.6483 - lr: 7.0117e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 145/750\n", - "\n", - "Epoch 145: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5843 - accuracy: 0.7333 - val_loss: 0.6542 - val_accuracy: 0.6483 - lr: 6.9942e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 146/750\n", - "\n", - "Epoch 146: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5841 - accuracy: 0.7333 - val_loss: 0.6542 - val_accuracy: 0.6483 - lr: 6.9767e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 147/750\n", - "\n", - "Epoch 147: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5839 - accuracy: 0.7333 - val_loss: 0.6541 - val_accuracy: 0.6483 - lr: 6.9593e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 148/750\n", - "\n", - "Epoch 148: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5837 - accuracy: 0.7345 - val_loss: 0.6540 - val_accuracy: 0.6483 - lr: 6.9419e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 149/750\n", - "\n", - "Epoch 149: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5835 - accuracy: 0.7333 - val_loss: 0.6538 - val_accuracy: 0.6483 - lr: 6.9246e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 150/750\n", - "\n", - "Epoch 150: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5833 - accuracy: 0.7333 - val_loss: 0.6537 - val_accuracy: 0.6483 - lr: 6.9073e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 151/750\n", - "\n", - "Epoch 151: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5831 - accuracy: 0.7333 - val_loss: 0.6535 - val_accuracy: 0.6483 - lr: 6.8901e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 152/750\n", - "\n", - "Epoch 152: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5829 - accuracy: 0.7333 - val_loss: 0.6535 - val_accuracy: 0.6483 - lr: 6.8729e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 153/750\n", - "\n", - "Epoch 153: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5827 - accuracy: 0.7333 - val_loss: 0.6534 - val_accuracy: 0.6483 - lr: 6.8557e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 154/750\n", - "\n", - "Epoch 154: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5825 - accuracy: 0.7333 - val_loss: 0.6534 - val_accuracy: 0.6483 - lr: 6.8386e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 155/750\n", - "\n", - "Epoch 155: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5823 - accuracy: 0.7345 - val_loss: 0.6534 - val_accuracy: 0.6483 - lr: 6.8215e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 156/750\n", - "\n", - "Epoch 156: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5821 - accuracy: 0.7345 - val_loss: 0.6533 - val_accuracy: 0.6483 - lr: 6.8045e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 157/750\n", - "\n", - "Epoch 157: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5820 - accuracy: 0.7345 - val_loss: 0.6533 - val_accuracy: 0.6483 - lr: 6.7875e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 158/750\n", - "\n", - "Epoch 158: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5818 - accuracy: 0.7369 - val_loss: 0.6534 - val_accuracy: 0.6483 - lr: 6.7705e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 159/750\n", - "\n", - "Epoch 159: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5818 - accuracy: 0.7369 - val_loss: 0.6534 - val_accuracy: 0.6483 - lr: 6.7536e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 160/750\n", - "\n", - "Epoch 160: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5816 - accuracy: 0.7369 - val_loss: 0.6532 - val_accuracy: 0.6483 - lr: 6.7368e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 161/750\n", - "\n", - "Epoch 161: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5814 - accuracy: 0.7369 - val_loss: 0.6531 - val_accuracy: 0.6483 - lr: 6.7200e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 162/750\n", - "\n", - "Epoch 162: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5812 - accuracy: 0.7369 - val_loss: 0.6530 - val_accuracy: 0.6483 - lr: 6.7032e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 163/750\n", - "\n", - "Epoch 163: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5810 - accuracy: 0.7369 - val_loss: 0.6528 - val_accuracy: 0.6483 - lr: 6.6864e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 164/750\n", - "\n", - "Epoch 164: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5808 - accuracy: 0.7345 - val_loss: 0.6526 - val_accuracy: 0.6483 - lr: 6.6697e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 165/750\n", - "\n", - "Epoch 165: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5806 - accuracy: 0.7345 - val_loss: 0.6525 - val_accuracy: 0.6483 - lr: 6.6531e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 166/750\n", - "\n", - "Epoch 166: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5804 - accuracy: 0.7345 - val_loss: 0.6524 - val_accuracy: 0.6483 - lr: 6.6365e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 167/750\n", - "\n", - "Epoch 167: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5803 - accuracy: 0.7333 - val_loss: 0.6522 - val_accuracy: 0.6483 - lr: 6.6199e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 168/750\n", - "\n", - "Epoch 168: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5801 - accuracy: 0.7320 - val_loss: 0.6522 - val_accuracy: 0.6483 - lr: 6.6034e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 169/750\n", - "\n", - "Epoch 169: val_accuracy did not improve from 0.64828\n", - "4/4 - 0s - loss: 0.5799 - accuracy: 0.7333 - val_loss: 0.6521 - val_accuracy: 0.6483 - lr: 6.5869e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 170/750\n", - "\n", - "Epoch 170: val_accuracy improved from 0.64828 to 0.65517, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.5797 - accuracy: 0.7357 - val_loss: 0.6520 - val_accuracy: 0.6552 - lr: 6.5704e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 171/750\n", - "\n", - "Epoch 171: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5796 - accuracy: 0.7369 - val_loss: 0.6519 - val_accuracy: 0.6483 - lr: 6.5540e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 172/750\n", - "\n", - "Epoch 172: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5794 - accuracy: 0.7333 - val_loss: 0.6517 - val_accuracy: 0.6483 - lr: 6.5377e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 173/750\n", - "\n", - "Epoch 173: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5792 - accuracy: 0.7320 - val_loss: 0.6516 - val_accuracy: 0.6483 - lr: 6.5214e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 174/750\n", - "\n", - "Epoch 174: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5790 - accuracy: 0.7333 - val_loss: 0.6515 - val_accuracy: 0.6483 - lr: 6.5051e-04 - 18ms/epoch - 5ms/step\n", - "Epoch 175/750\n", - "\n", - "Epoch 175: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5788 - accuracy: 0.7333 - val_loss: 0.6514 - val_accuracy: 0.6483 - lr: 6.4888e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 176/750\n", - "\n", - "Epoch 176: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5787 - accuracy: 0.7345 - val_loss: 0.6514 - val_accuracy: 0.6483 - lr: 6.4726e-04 - 18ms/epoch - 4ms/step\n", - "Epoch 177/750\n", - "\n", - "Epoch 177: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5785 - accuracy: 0.7345 - val_loss: 0.6513 - val_accuracy: 0.6483 - lr: 6.4565e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 178/750\n", - "\n", - "Epoch 178: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5783 - accuracy: 0.7345 - val_loss: 0.6513 - val_accuracy: 0.6483 - lr: 6.4403e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 179/750\n", - "\n", - "Epoch 179: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5782 - accuracy: 0.7345 - val_loss: 0.6512 - val_accuracy: 0.6483 - lr: 6.4243e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 180/750\n", - "\n", - "Epoch 180: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5781 - accuracy: 0.7345 - val_loss: 0.6511 - val_accuracy: 0.6483 - lr: 6.4082e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 181/750\n", - "\n", - "Epoch 181: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5778 - accuracy: 0.7357 - val_loss: 0.6510 - val_accuracy: 0.6552 - lr: 6.3922e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 182/750\n", - "\n", - "Epoch 182: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5776 - accuracy: 0.7357 - val_loss: 0.6509 - val_accuracy: 0.6552 - lr: 6.3763e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 183/750\n", - "\n", - "Epoch 183: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5775 - accuracy: 0.7369 - val_loss: 0.6507 - val_accuracy: 0.6552 - lr: 6.3603e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 184/750\n", - "\n", - "Epoch 184: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5773 - accuracy: 0.7369 - val_loss: 0.6506 - val_accuracy: 0.6552 - lr: 6.3445e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 185/750\n", - "\n", - "Epoch 185: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5771 - accuracy: 0.7357 - val_loss: 0.6505 - val_accuracy: 0.6552 - lr: 6.3286e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 186/750\n", - "\n", - "Epoch 186: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5770 - accuracy: 0.7369 - val_loss: 0.6504 - val_accuracy: 0.6552 - lr: 6.3128e-04 - 17ms/epoch - 4ms/step\n", - "Epoch 187/750\n", - "\n", - "Epoch 187: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5768 - accuracy: 0.7369 - val_loss: 0.6503 - val_accuracy: 0.6552 - lr: 6.2971e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 188/750\n", - "\n", - "Epoch 188: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5767 - accuracy: 0.7369 - val_loss: 0.6502 - val_accuracy: 0.6552 - lr: 6.2813e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 189/750\n", - "\n", - "Epoch 189: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5765 - accuracy: 0.7369 - val_loss: 0.6501 - val_accuracy: 0.6552 - lr: 6.2656e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 190/750\n", - "\n", - "Epoch 190: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5763 - accuracy: 0.7357 - val_loss: 0.6500 - val_accuracy: 0.6552 - lr: 6.2500e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 191/750\n", - "\n", - "Epoch 191: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5763 - accuracy: 0.7345 - val_loss: 0.6500 - val_accuracy: 0.6552 - lr: 6.2344e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 192/750\n", - "\n", - "Epoch 192: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5760 - accuracy: 0.7357 - val_loss: 0.6499 - val_accuracy: 0.6552 - lr: 6.2188e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 193/750\n", - "\n", - "Epoch 193: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5758 - accuracy: 0.7357 - val_loss: 0.6498 - val_accuracy: 0.6552 - lr: 6.2033e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 194/750\n", - "\n", - "Epoch 194: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5757 - accuracy: 0.7357 - val_loss: 0.6498 - val_accuracy: 0.6552 - lr: 6.1878e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 195/750\n", - "\n", - "Epoch 195: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5756 - accuracy: 0.7369 - val_loss: 0.6497 - val_accuracy: 0.6552 - lr: 6.1724e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 196/750\n", - "\n", - "Epoch 196: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5754 - accuracy: 0.7369 - val_loss: 0.6497 - val_accuracy: 0.6552 - lr: 6.1570e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 197/750\n", - "\n", - "Epoch 197: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5753 - accuracy: 0.7369 - val_loss: 0.6498 - val_accuracy: 0.6552 - lr: 6.1416e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 198/750\n", - "\n", - "Epoch 198: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5752 - accuracy: 0.7369 - val_loss: 0.6497 - val_accuracy: 0.6552 - lr: 6.1262e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 199/750\n", - "\n", - "Epoch 199: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5750 - accuracy: 0.7369 - val_loss: 0.6497 - val_accuracy: 0.6552 - lr: 6.1109e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 200/750\n", - "\n", - "Epoch 200: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5748 - accuracy: 0.7381 - val_loss: 0.6496 - val_accuracy: 0.6552 - lr: 6.0957e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 201/750\n", - "\n", - "Epoch 201: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5747 - accuracy: 0.7381 - val_loss: 0.6496 - val_accuracy: 0.6552 - lr: 6.0805e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 202/750\n", - "\n", - "Epoch 202: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5746 - accuracy: 0.7381 - val_loss: 0.6496 - val_accuracy: 0.6552 - lr: 6.0653e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 203/750\n", - "\n", - "Epoch 203: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5744 - accuracy: 0.7381 - val_loss: 0.6496 - val_accuracy: 0.6552 - lr: 6.0501e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 204/750\n", - "\n", - "Epoch 204: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5742 - accuracy: 0.7381 - val_loss: 0.6495 - val_accuracy: 0.6552 - lr: 6.0350e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 205/750\n", - "\n", - "Epoch 205: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5741 - accuracy: 0.7393 - val_loss: 0.6493 - val_accuracy: 0.6552 - lr: 6.0200e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 206/750\n", - "\n", - "Epoch 206: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5740 - accuracy: 0.7418 - val_loss: 0.6492 - val_accuracy: 0.6552 - lr: 6.0049e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 207/750\n", - "\n", - "Epoch 207: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5738 - accuracy: 0.7406 - val_loss: 0.6492 - val_accuracy: 0.6552 - lr: 5.9899e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 208/750\n", - "\n", - "Epoch 208: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5737 - accuracy: 0.7418 - val_loss: 0.6492 - val_accuracy: 0.6552 - lr: 5.9750e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 209/750\n", - "\n", - "Epoch 209: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5736 - accuracy: 0.7418 - val_loss: 0.6491 - val_accuracy: 0.6552 - lr: 5.9601e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 210/750\n", - "\n", - "Epoch 210: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5734 - accuracy: 0.7406 - val_loss: 0.6490 - val_accuracy: 0.6552 - lr: 5.9452e-04 - 34ms/epoch - 8ms/step\n", - "Epoch 211/750\n", - "\n", - "Epoch 211: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5733 - accuracy: 0.7393 - val_loss: 0.6489 - val_accuracy: 0.6552 - lr: 5.9303e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 212/750\n", - "\n", - "Epoch 212: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5732 - accuracy: 0.7393 - val_loss: 0.6488 - val_accuracy: 0.6552 - lr: 5.9155e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 213/750\n", - "\n", - "Epoch 213: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5731 - accuracy: 0.7406 - val_loss: 0.6487 - val_accuracy: 0.6552 - lr: 5.9008e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 214/750\n", - "\n", - "Epoch 214: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5730 - accuracy: 0.7393 - val_loss: 0.6487 - val_accuracy: 0.6552 - lr: 5.8860e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 215/750\n", - "\n", - "Epoch 215: val_accuracy did not improve from 0.65517\n", - "4/4 - 0s - loss: 0.5728 - accuracy: 0.7393 - val_loss: 0.6486 - val_accuracy: 0.6552 - lr: 5.8713e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 216/750\n", - "\n", - "Epoch 216: val_accuracy improved from 0.65517 to 0.66207, saving model to best_model.h5\n", - "4/4 - 0s - loss: 0.5726 - accuracy: 0.7393 - val_loss: 0.6484 - val_accuracy: 0.6621 - lr: 5.8567e-04 - 21ms/epoch - 5ms/step\n", - "Epoch 217/750\n", - "\n", - "Epoch 217: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5725 - accuracy: 0.7393 - val_loss: 0.6483 - val_accuracy: 0.6621 - lr: 5.8420e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 218/750\n", - "\n", - "Epoch 218: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5724 - accuracy: 0.7381 - val_loss: 0.6482 - val_accuracy: 0.6621 - lr: 5.8275e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 219/750\n", - "\n", - "Epoch 219: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5723 - accuracy: 0.7393 - val_loss: 0.6482 - val_accuracy: 0.6621 - lr: 5.8129e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 220/750\n", - "\n", - "Epoch 220: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5721 - accuracy: 0.7393 - val_loss: 0.6482 - val_accuracy: 0.6621 - lr: 5.7984e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 221/750\n", - "\n", - "Epoch 221: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5720 - accuracy: 0.7393 - val_loss: 0.6482 - val_accuracy: 0.6621 - lr: 5.7839e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 222/750\n", - "\n", - "Epoch 222: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5719 - accuracy: 0.7393 - val_loss: 0.6483 - val_accuracy: 0.6621 - lr: 5.7695e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 223/750\n", - "\n", - "Epoch 223: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5718 - accuracy: 0.7393 - val_loss: 0.6485 - val_accuracy: 0.6552 - lr: 5.7551e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 224/750\n", - "\n", - "Epoch 224: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5718 - accuracy: 0.7393 - val_loss: 0.6485 - val_accuracy: 0.6552 - lr: 5.7407e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 225/750\n", - "\n", - "Epoch 225: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5717 - accuracy: 0.7393 - val_loss: 0.6485 - val_accuracy: 0.6552 - lr: 5.7264e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 226/750\n", - "\n", - "Epoch 226: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5715 - accuracy: 0.7393 - val_loss: 0.6485 - val_accuracy: 0.6552 - lr: 5.7121e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 227/750\n", - "\n", - "Epoch 227: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5714 - accuracy: 0.7393 - val_loss: 0.6486 - val_accuracy: 0.6552 - lr: 5.6978e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 228/750\n", - "\n", - "Epoch 228: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5712 - accuracy: 0.7393 - val_loss: 0.6486 - val_accuracy: 0.6552 - lr: 5.6836e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 229/750\n", - "\n", - "Epoch 229: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5712 - accuracy: 0.7393 - val_loss: 0.6486 - val_accuracy: 0.6552 - lr: 5.6694e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 230/750\n", - "\n", - "Epoch 230: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5710 - accuracy: 0.7393 - val_loss: 0.6487 - val_accuracy: 0.6552 - lr: 5.6552e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 231/750\n", - "\n", - "Epoch 231: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5710 - accuracy: 0.7406 - val_loss: 0.6487 - val_accuracy: 0.6552 - lr: 5.6411e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 232/750\n", - "\n", - "Epoch 232: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5708 - accuracy: 0.7418 - val_loss: 0.6487 - val_accuracy: 0.6621 - lr: 5.6270e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 233/750\n", - "\n", - "Epoch 233: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5707 - accuracy: 0.7406 - val_loss: 0.6485 - val_accuracy: 0.6621 - lr: 5.6130e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 234/750\n", - "\n", - "Epoch 234: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5705 - accuracy: 0.7393 - val_loss: 0.6485 - val_accuracy: 0.6621 - lr: 5.5990e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 235/750\n", - "\n", - "Epoch 235: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5705 - accuracy: 0.7381 - val_loss: 0.6485 - val_accuracy: 0.6621 - lr: 5.5850e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 236/750\n", - "\n", - "Epoch 236: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5703 - accuracy: 0.7406 - val_loss: 0.6485 - val_accuracy: 0.6621 - lr: 5.5710e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 237/750\n", - "\n", - "Epoch 237: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5703 - accuracy: 0.7393 - val_loss: 0.6485 - val_accuracy: 0.6621 - lr: 5.5571e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 238/750\n", - "\n", - "Epoch 238: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5701 - accuracy: 0.7406 - val_loss: 0.6484 - val_accuracy: 0.6621 - lr: 5.5432e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 239/750\n", - "\n", - "Epoch 239: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5700 - accuracy: 0.7406 - val_loss: 0.6484 - val_accuracy: 0.6621 - lr: 5.5294e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 240/750\n", - "\n", - "Epoch 240: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5698 - accuracy: 0.7406 - val_loss: 0.6483 - val_accuracy: 0.6621 - lr: 5.5156e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 241/750\n", - "\n", - "Epoch 241: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5697 - accuracy: 0.7406 - val_loss: 0.6483 - val_accuracy: 0.6621 - lr: 5.5018e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 242/750\n", - "\n", - "Epoch 242: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5696 - accuracy: 0.7381 - val_loss: 0.6483 - val_accuracy: 0.6621 - lr: 5.4881e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 243/750\n", - "\n", - "Epoch 243: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5695 - accuracy: 0.7393 - val_loss: 0.6482 - val_accuracy: 0.6621 - lr: 5.4744e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 244/750\n", - "\n", - "Epoch 244: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5694 - accuracy: 0.7393 - val_loss: 0.6481 - val_accuracy: 0.6621 - lr: 5.4607e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 245/750\n", - "\n", - "Epoch 245: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5692 - accuracy: 0.7406 - val_loss: 0.6481 - val_accuracy: 0.6621 - lr: 5.4471e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 246/750\n", - "\n", - "Epoch 246: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5691 - accuracy: 0.7406 - val_loss: 0.6480 - val_accuracy: 0.6621 - lr: 5.4335e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 247/750\n", - "\n", - "Epoch 247: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5690 - accuracy: 0.7406 - val_loss: 0.6480 - val_accuracy: 0.6621 - lr: 5.4199e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 248/750\n", - "\n", - "Epoch 248: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5689 - accuracy: 0.7406 - val_loss: 0.6480 - val_accuracy: 0.6621 - lr: 5.4064e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 249/750\n", - "\n", - "Epoch 249: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5688 - accuracy: 0.7406 - val_loss: 0.6480 - val_accuracy: 0.6621 - lr: 5.3929e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 250/750\n", - "\n", - "Epoch 250: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5687 - accuracy: 0.7418 - val_loss: 0.6480 - val_accuracy: 0.6621 - lr: 5.3794e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 251/750\n", - "\n", - "Epoch 251: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5686 - accuracy: 0.7418 - val_loss: 0.6479 - val_accuracy: 0.6621 - lr: 5.3660e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 252/750\n", - "\n", - "Epoch 252: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5685 - accuracy: 0.7406 - val_loss: 0.6479 - val_accuracy: 0.6621 - lr: 5.3526e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 253/750\n", - "\n", - "Epoch 253: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5684 - accuracy: 0.7393 - val_loss: 0.6478 - val_accuracy: 0.6621 - lr: 5.3392e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 254/750\n", - "\n", - "Epoch 254: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5683 - accuracy: 0.7393 - val_loss: 0.6477 - val_accuracy: 0.6621 - lr: 5.3259e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 255/750\n", - "\n", - "Epoch 255: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5682 - accuracy: 0.7406 - val_loss: 0.6477 - val_accuracy: 0.6621 - lr: 5.3126e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 256/750\n", - "\n", - "Epoch 256: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5681 - accuracy: 0.7406 - val_loss: 0.6476 - val_accuracy: 0.6621 - lr: 5.2993e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 257/750\n", - "\n", - "Epoch 257: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5680 - accuracy: 0.7406 - val_loss: 0.6476 - val_accuracy: 0.6621 - lr: 5.2861e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 258/750\n", - "\n", - "Epoch 258: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5679 - accuracy: 0.7406 - val_loss: 0.6475 - val_accuracy: 0.6621 - lr: 5.2729e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 259/750\n", - "\n", - "Epoch 259: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5678 - accuracy: 0.7418 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 5.2597e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 260/750\n", - "\n", - "Epoch 260: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5677 - accuracy: 0.7418 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 5.2466e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 261/750\n", - "\n", - "Epoch 261: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5676 - accuracy: 0.7406 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 5.2335e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 262/750\n", - "\n", - "Epoch 262: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5675 - accuracy: 0.7418 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 5.2204e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 263/750\n", - "\n", - "Epoch 263: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5674 - accuracy: 0.7418 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 5.2074e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 264/750\n", - "\n", - "Epoch 264: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5673 - accuracy: 0.7418 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 5.1944e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 265/750\n", - "\n", - "Epoch 265: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5672 - accuracy: 0.7418 - val_loss: 0.6473 - val_accuracy: 0.6621 - lr: 5.1814e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 266/750\n", - "\n", - "Epoch 266: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5671 - accuracy: 0.7418 - val_loss: 0.6473 - val_accuracy: 0.6621 - lr: 5.1685e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 267/750\n", - "\n", - "Epoch 267: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5670 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.1556e-04 - 16ms/epoch - 4ms/step\n", - "Epoch 268/750\n", - "\n", - "Epoch 268: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5669 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.1427e-04 - 15ms/epoch - 4ms/step\n", - "Epoch 269/750\n", - "\n", - "Epoch 269: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5668 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.1299e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 270/750\n", - "\n", - "Epoch 270: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5666 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.1171e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 271/750\n", - "\n", - "Epoch 271: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5666 - accuracy: 0.7430 - val_loss: 0.6473 - val_accuracy: 0.6621 - lr: 5.1043e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 272/750\n", - "\n", - "Epoch 272: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5665 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.0915e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 273/750\n", - "\n", - "Epoch 273: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5664 - accuracy: 0.7430 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.0788e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 274/750\n", - "\n", - "Epoch 274: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5663 - accuracy: 0.7430 - val_loss: 0.6471 - val_accuracy: 0.6621 - lr: 5.0661e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 275/750\n", - "\n", - "Epoch 275: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5662 - accuracy: 0.7430 - val_loss: 0.6471 - val_accuracy: 0.6621 - lr: 5.0535e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 276/750\n", - "\n", - "Epoch 276: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5660 - accuracy: 0.7418 - val_loss: 0.6471 - val_accuracy: 0.6621 - lr: 5.0409e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 277/750\n", - "\n", - "Epoch 277: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5660 - accuracy: 0.7430 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.0283e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 278/750\n", - "\n", - "Epoch 278: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5659 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 5.0157e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 279/750\n", - "\n", - "Epoch 279: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5657 - accuracy: 0.7418 - val_loss: 0.6473 - val_accuracy: 0.6621 - lr: 5.0032e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 280/750\n", - "\n", - "Epoch 280: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5657 - accuracy: 0.7418 - val_loss: 0.6473 - val_accuracy: 0.6621 - lr: 4.9907e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 281/750\n", - "\n", - "Epoch 281: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5656 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 4.9783e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 282/750\n", - "\n", - "Epoch 282: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5654 - accuracy: 0.7418 - val_loss: 0.6471 - val_accuracy: 0.6621 - lr: 4.9658e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 283/750\n", - "\n", - "Epoch 283: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5655 - accuracy: 0.7418 - val_loss: 0.6470 - val_accuracy: 0.6621 - lr: 4.9534e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 284/750\n", - "\n", - "Epoch 284: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5654 - accuracy: 0.7430 - val_loss: 0.6469 - val_accuracy: 0.6621 - lr: 4.9411e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 285/750\n", - "\n", - "Epoch 285: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5654 - accuracy: 0.7406 - val_loss: 0.6468 - val_accuracy: 0.6621 - lr: 4.9287e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 286/750\n", - "\n", - "Epoch 286: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5653 - accuracy: 0.7430 - val_loss: 0.6466 - val_accuracy: 0.6621 - lr: 4.9164e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 287/750\n", - "\n", - "Epoch 287: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5651 - accuracy: 0.7418 - val_loss: 0.6465 - val_accuracy: 0.6621 - lr: 4.9041e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 288/750\n", - "\n", - "Epoch 288: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5650 - accuracy: 0.7430 - val_loss: 0.6464 - val_accuracy: 0.6621 - lr: 4.8919e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 289/750\n", - "\n", - "Epoch 289: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5649 - accuracy: 0.7430 - val_loss: 0.6463 - val_accuracy: 0.6621 - lr: 4.8797e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 290/750\n", - "\n", - "Epoch 290: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5648 - accuracy: 0.7430 - val_loss: 0.6464 - val_accuracy: 0.6621 - lr: 4.8675e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 291/750\n", - "\n", - "Epoch 291: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5647 - accuracy: 0.7430 - val_loss: 0.6464 - val_accuracy: 0.6621 - lr: 4.8553e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 292/750\n", - "\n", - "Epoch 292: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5646 - accuracy: 0.7418 - val_loss: 0.6464 - val_accuracy: 0.6621 - lr: 4.8432e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 293/750\n", - "\n", - "Epoch 293: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5645 - accuracy: 0.7418 - val_loss: 0.6465 - val_accuracy: 0.6621 - lr: 4.8311e-04 - 38ms/epoch - 9ms/step\n", - "Epoch 294/750\n", - "\n", - "Epoch 294: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5644 - accuracy: 0.7418 - val_loss: 0.6465 - val_accuracy: 0.6621 - lr: 4.8191e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 295/750\n", - "\n", - "Epoch 295: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5643 - accuracy: 0.7418 - val_loss: 0.6465 - val_accuracy: 0.6621 - lr: 4.8070e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 296/750\n", - "\n", - "Epoch 296: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5642 - accuracy: 0.7418 - val_loss: 0.6465 - val_accuracy: 0.6621 - lr: 4.7950e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 297/750\n", - "\n", - "Epoch 297: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5641 - accuracy: 0.7418 - val_loss: 0.6465 - val_accuracy: 0.6621 - lr: 4.7831e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 298/750\n", - "\n", - "Epoch 298: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5640 - accuracy: 0.7418 - val_loss: 0.6466 - val_accuracy: 0.6621 - lr: 4.7711e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 299/750\n", - "\n", - "Epoch 299: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5639 - accuracy: 0.7418 - val_loss: 0.6466 - val_accuracy: 0.6621 - lr: 4.7592e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 300/750\n", - "\n", - "Epoch 300: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5638 - accuracy: 0.7442 - val_loss: 0.6467 - val_accuracy: 0.6621 - lr: 4.7473e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 301/750\n", - "\n", - "Epoch 301: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5637 - accuracy: 0.7442 - val_loss: 0.6468 - val_accuracy: 0.6621 - lr: 4.7355e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 302/750\n", - "\n", - "Epoch 302: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5636 - accuracy: 0.7418 - val_loss: 0.6468 - val_accuracy: 0.6621 - lr: 4.7236e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 303/750\n", - "\n", - "Epoch 303: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5635 - accuracy: 0.7418 - val_loss: 0.6468 - val_accuracy: 0.6621 - lr: 4.7118e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 304/750\n", - "\n", - "Epoch 304: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5634 - accuracy: 0.7406 - val_loss: 0.6468 - val_accuracy: 0.6621 - lr: 4.7001e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 305/750\n", - "\n", - "Epoch 305: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5633 - accuracy: 0.7406 - val_loss: 0.6469 - val_accuracy: 0.6621 - lr: 4.6883e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 306/750\n", - "\n", - "Epoch 306: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5632 - accuracy: 0.7406 - val_loss: 0.6470 - val_accuracy: 0.6621 - lr: 4.6766e-04 - 14ms/epoch - 3ms/step\n", - "Epoch 307/750\n", - "\n", - "Epoch 307: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5631 - accuracy: 0.7418 - val_loss: 0.6471 - val_accuracy: 0.6621 - lr: 4.6650e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 308/750\n", - "\n", - "Epoch 308: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5630 - accuracy: 0.7418 - val_loss: 0.6472 - val_accuracy: 0.6621 - lr: 4.6533e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 309/750\n", - "\n", - "Epoch 309: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5629 - accuracy: 0.7418 - val_loss: 0.6473 - val_accuracy: 0.6621 - lr: 4.6417e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 310/750\n", - "\n", - "Epoch 310: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5628 - accuracy: 0.7418 - val_loss: 0.6474 - val_accuracy: 0.6621 - lr: 4.6301e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 311/750\n", - "\n", - "Epoch 311: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5627 - accuracy: 0.7418 - val_loss: 0.6475 - val_accuracy: 0.6621 - lr: 4.6185e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 312/750\n", - "\n", - "Epoch 312: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5627 - accuracy: 0.7418 - val_loss: 0.6476 - val_accuracy: 0.6621 - lr: 4.6070e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 313/750\n", - "\n", - "Epoch 313: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5626 - accuracy: 0.7418 - val_loss: 0.6477 - val_accuracy: 0.6621 - lr: 4.5955e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 314/750\n", - "\n", - "Epoch 314: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5625 - accuracy: 0.7418 - val_loss: 0.6477 - val_accuracy: 0.6621 - lr: 4.5840e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 315/750\n", - "\n", - "Epoch 315: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5625 - accuracy: 0.7406 - val_loss: 0.6477 - val_accuracy: 0.6621 - lr: 4.5726e-04 - 14ms/epoch - 4ms/step\n", - "Epoch 316/750\n", - "\n", - "Epoch 316: val_accuracy did not improve from 0.66207\n", - "4/4 - 0s - loss: 0.5623 - accuracy: 0.7406 - val_loss: 0.6477 - val_accuracy: 0.6621 - lr: 4.5612e-04 - 14ms/epoch - 4ms/step\n" - ] - } - ], + "outputs": [], "source": [ "BATCH_SIZE = int(len(y)/6.6125) #128 #24 #4\n", "BATCH_SIZE = 256\n", @@ -1870,30 +317,9 @@ }, { "cell_type": "code", - "execution_count": 121, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], + "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "\n", @@ -1962,7 +388,7 @@ }, { "cell_type": "code", - "execution_count": 124, + "execution_count": 26, "metadata": {}, "outputs": [], "source": [ @@ -1973,65 +399,33 @@ }, { "cell_type": "code", - "execution_count": 126, + "execution_count": 27, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - "07-09-2023 Nifty Prediction -> Market may Close BULLISH on 08-09-2023! Actual -> BULLISH, Prediction -> Correct, Pred = nan\n", - "08-09-2023 Nifty Prediction -> Market may Close BULLISH on 09-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "11-09-2023 Nifty Prediction -> Market may Close BULLISH on 12-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "12-09-2023 Nifty Prediction -> Market may Close BULLISH on 13-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "13-09-2023 Nifty Prediction -> Market may Close BULLISH on 14-09-2023! Actual -> BULLISH, Prediction -> Correct, Pred = nan\n", - "14-09-2023 Nifty Prediction -> Market may Close BULLISH on 15-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "15-09-2023 Nifty Prediction -> Market may Close BULLISH on 16-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "18-09-2023 Nifty Prediction -> Market may Close BULLISH on 19-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "20-09-2023 Nifty Prediction -> Market may Close BULLISH on 21-09-2023! Actual -> BULLISH, Prediction -> Correct, Pred = nan\n", - "21-09-2023 Nifty Prediction -> Market may Close BEARISH on 22-09-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "22-09-2023 Nifty Prediction -> Market may Close BEARISH on 23-09-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "25-09-2023 Nifty Prediction -> Market may Close BULLISH on 26-09-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.5\n", - "26-09-2023 Nifty Prediction -> Market may Close BEARISH on 27-09-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.58\n", - "27-09-2023 Nifty Prediction -> Market may Close BULLISH on 28-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.27\n", - "28-09-2023 Nifty Prediction -> Market may Close BEARISH on 29-09-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "29-09-2023 Nifty Prediction -> Market may Close BULLISH on 30-09-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.22\n", - "02-10-2023 Nifty Prediction -> Market may Close BULLISH on 03-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "03-10-2023 Nifty Prediction -> Market may Close BULLISH on 04-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "04-10-2023 Nifty Prediction -> Market may Close BEARISH on 05-10-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "05-10-2023 Nifty Prediction -> Market may Close BULLISH on 06-10-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.29\n", - "06-10-2023 Nifty Prediction -> Market may Close BULLISH on 07-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.33\n", - "09-10-2023 Nifty Prediction -> Market may Close BEARISH on 10-10-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "10-10-2023 Nifty Prediction -> Market may Close BULLISH on 11-10-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.2\n", - "11-10-2023 Nifty Prediction -> Market may Close BULLISH on 12-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.44\n", - "12-10-2023 Nifty Prediction -> Market may Close BEARISH on 13-10-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "13-10-2023 Nifty Prediction -> Market may Close BULLISH on 14-10-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.18\n", - "16-10-2023 Nifty Prediction -> Market may Close BEARISH on 17-10-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "17-10-2023 Nifty Prediction -> Market may Close BULLISH on 18-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.39\n", - "18-10-2023 Nifty Prediction -> Market may Close BEARISH on 19-10-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.53\n", - "19-10-2023 Nifty Prediction -> Market may Close BULLISH on 20-10-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.47\n", - "20-10-2023 Nifty Prediction -> Market may Close BEARISH on 21-10-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "23-10-2023 Nifty Prediction -> Market may Close BEARISH on 24-10-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "24-10-2023 Nifty Prediction -> Market may Close BULLISH on 25-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "25-10-2023 Nifty Prediction -> Market may Close BULLISH on 26-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "26-10-2023 Nifty Prediction -> Market may Close BEARISH on 27-10-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "27-10-2023 Nifty Prediction -> Market may Close BULLISH on 28-10-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.23\n", - "30-10-2023 Nifty Prediction -> Market may Close BULLISH on 31-10-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.34\n", - "31-10-2023 Nifty Prediction -> Market may Close BEARISH on 01-11-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "01-11-2023 Nifty Prediction -> Market may Close BEARISH on 02-11-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "02-11-2023 Nifty Prediction -> Market may Close BULLISH on 03-11-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.3\n", - "03-11-2023 Nifty Prediction -> Market may Close BULLISH on 04-11-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.39\n", - "06-11-2023 Nifty Prediction -> Market may Close BULLISH on 07-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.21\n", - "07-11-2023 Nifty Prediction -> Market may Close BEARISH on 08-11-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "08-11-2023 Nifty Prediction -> Market may Close BULLISH on 09-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.42\n", - "09-11-2023 Nifty Prediction -> Market may Close BEARISH on 10-11-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", - "10-11-2023 Nifty Prediction -> Market may Close BULLISH on 11-11-2023! Actual -> BULLISH, Prediction -> Correct, Pred = 0.28\n", - "13-11-2023 Nifty Prediction -> Market may Close BEARISH on 14-11-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", - "14-11-2023 Nifty Prediction -> Market may Close BULLISH on 15-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "15-11-2023 Nifty Prediction -> Market may Close BULLISH on 16-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = nan\n", - "16-11-2023 Nifty Prediction -> Market may Close BULLISH on 17-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.26\n", - "Correct: 21, Wrong: 29, Accuracy: 0.42\n", - "{'TP': 12, 'FP': 8, 'TN': 9, 'FN': 21}\n" + " Open High Low Close gold_Close crude_Close\n", + "Date \n", + "2023-11-15 0.697093 0.221577 0.441300 0.093698 -0.086659 -2.044465\n", + "2023-11-16 0.118561 0.924435 0.241831 0.456152 1.214226 -4.904777\n", + "2023-11-17 0.000258 -0.348423 0.206090 -0.168976 -0.115936 4.101506\n", + "2023-11-20 0.286664 -0.250181 0.015512 -0.191573 -0.196812 2.253260\n", + "2023-11-21 0.201458 0.367730 0.424752 0.453947 1.092183 0.219070\n", + "2023-11-22 0.066257 -0.017897 -0.254131 0.143803 -0.395140 -0.861512\n", + "2023-11-23 0.224673 0.250180 0.420732 -0.049716 -0.395140 -0.861512\n", + "2023-11-24 -0.095063 -0.212833 -0.090467 -0.036869 -0.395140 -0.861512\n", + "15-11-2023 Nifty Prediction -> Market may Close BEARISH on 16-11-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", + "16-11-2023 Nifty Prediction -> Market may Close BULLISH on 17-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.2\n", + "17-11-2023 Nifty Prediction -> Market may Close BEARISH on 18-11-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.59\n", + "20-11-2023 Nifty Prediction -> Market may Close BEARISH on 21-11-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.59\n", + "21-11-2023 Nifty Prediction -> Market may Close BULLISH on 22-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.28\n", + "22-11-2023 Nifty Prediction -> Market may Close BULLISH on 23-11-2023! Actual -> BEARISH, Prediction -> Wrong, Pred = 0.48\n", + "23-11-2023 Nifty Prediction -> Market may Close BEARISH on 24-11-2023! Actual -> BEARISH, Prediction -> Correct, Pred = 0.57\n", + "24-11-2023 Nifty Prediction -> Market may Close BEARISH on 25-11-2023! Actual -> BULLISH, Prediction -> Wrong, Pred = 0.56\n", + "Correct: 3, Wrong: 5, Accuracy: 0.375\n", + "{'TP': 0, 'FP': 2, 'TN': 3, 'FN': 3}\n" ] } ], @@ -2041,7 +435,8 @@ "try:\n", " scaler\n", "except NameError:\n", - " pkl = joblib.load('nifty_model.pkl')\n", + " # pkl = joblib.load('nifty_model.pkl')\n", + " pkl = joblib.load('nifty_model_v3.pkl')\n", " scaler = pkl['scaler']\n", "today = yf.download(\n", " tickers=\"^NSEI\",\n", @@ -2074,12 +469,16 @@ "###\n", "today = preprocessBeforeScaling(today)\n", "today = today.drop(columns=['gold_Open', 'gold_High', 'gold_Low', 'crude_Open', 'crude_High', 'crude_Low'])\n", + "print(today)\n", "###\n", "\n", "cnt_correct, cnt_wrong = 0, 0\n", "for i in range(-TEST_DAYS,0):\n", - " df = today.iloc[i]\n", - " twr = today.iloc[i+1]['Close']\n", + " try:\n", + " df = today.iloc[i]\n", + " twr = today.iloc[i+1]['Close']\n", + " except IndexError:\n", + " continue\n", " df = scaler.transform([df])\n", " pred = endpoint.predict([df], verbose=0)\n", "\n", @@ -2130,20 +529,9 @@ }, { "cell_type": "code", - "execution_count": 127, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "data": { - "text/plain": [ - "['nifty_model.pkl']" - ] - }, - "execution_count": 127, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "pkl = {\n", " # 'model': model,\n", @@ -2156,34 +544,9 @@ }, { "cell_type": "code", - "execution_count": 128, + "execution_count": null, "metadata": {}, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Open 0.118561\n", - "Close 0.646497\n", - "High 0.608594\n", - "Low 0.241831\n", - "gold_Close 0.397941\n", - "crude_Close -1.174017\n", - "Name: 2023-11-16 00:00:00, dtype: float64\n", - "1/1 [==============================] - 0s 40ms/step\n" - ] - }, - { - "data": { - "text/plain": [ - "array([[0.4365444]], dtype=float32)" - ] - }, - "execution_count": 128, - "metadata": {}, - "output_type": "execute_result" - } - ], + "outputs": [], "source": [ "pkl = joblib.load('nifty_model_v3.pkl')\n", "z = yf.download(\n", diff --git a/src/screenipy.py b/src/screenipy.py index f97043d4..7957f913 100644 --- a/src/screenipy.py +++ b/src/screenipy.py @@ -204,6 +204,9 @@ def main(testing=False, testBuild=False, downloadOnly=False, execute_inputs:list configManager.setConfig(ConfigManager.parser, default=True, showFileCreatedText=False) try: tickerOption, executeOption = int(execute_inputs[0]), int(execute_inputs[1]) + if tickerOption == 0: + stockCode = execute_inputs[2].replace(" ", "") + listStockCodes = stockCode.split(',') except: tickerOption, executeOption = str(execute_inputs[0]), int(execute_inputs[1]) if tickerOption == 13: @@ -363,7 +366,8 @@ def main(testing=False, testBuild=False, downloadOnly=False, execute_inputs:list configManager.stageTwo = False configManager.minLTP = 0.1 configManager.maxLTP = 999999999 - listStockCodes = fetcher.fetchStockCodes(tickerOption, proxyServer=proxyServer) + if (execute_inputs != [] and tickerOption != 0) or execute_inputs == []: + listStockCodes = fetcher.fetchStockCodes(tickerOption, proxyServer=proxyServer) except urllib.error.URLError: print(colorText.BOLD + colorText.FAIL + "\n\n[+] Oops! It looks like you don't have an Internet connectivity at the moment! Press any key to exit!" + colorText.END) diff --git a/src/streamlit_app.py b/src/streamlit_app.py index 2e124984..9425f0ea 100644 --- a/src/streamlit_app.py +++ b/src/streamlit_app.py @@ -175,11 +175,8 @@ def get_extra_inputs(tickerOption, executeOption, c_index=None, c_criteria=None, if not tickerOption.isnumeric(): execute_inputs = [tickerOption, 0, 'N'] elif int(tickerOption) == 0 or tickerOption is None: - stock_codes = c_index.text_input('Enter Stock Code(s) (Multiple codes should be seperated by ,)', placeholder='SBIN, INFY, ITC') - if stock_codes: - execute_inputs = [tickerOption, executeOption, stock_codes, 'N'] - else: - c_index.error("Stock codes can't be left blank!") + stock_codes = c_index.text_input('Enter Stock Code(s)', placeholder='SBIN, INFY, ITC') + execute_inputs = [tickerOption, executeOption, stock_codes, 'N'] elif int(executeOption) >= 0 and int(executeOption) < 4: execute_inputs = [tickerOption, executeOption, 'N'] elif int(executeOption) == 4: @@ -287,7 +284,7 @@ def get_extra_inputs(tickerOption, executeOption, c_index=None, c_criteria=None, # 'W > Screen stocks from my own Watchlist', # 'N > Nifty Prediction using Artifical Intelligence (Use for Gap-Up/Gap-Down/BTST/STBT)', # 'E > Live Index Scan : 5 EMA for Intraday', - # '0 > Screen stocks by the stock names (NSE Stock Code)', + '0 > By Stock Names (NSE Stock Code)', '1 > Nifty 50', '2 > Nifty Next 50', '3 > Nifty 100',