This is a sample implementation of RNN(Recurrent Neural Networks) using LSTM layer.
The dataset used is mnist dataset.
- Clone and download the project
git clone https://github.com/jay24rajput/Mnist-Digit-Recognision.git
- Install the dependencies
pip3 install -r requirements.txt --user
- Run the python file
python3 Digit_Recognision.py
If you are running the program on CPU then make the following changes:
line 3
from tensorflow.keras.layers import Dense, Dropout,LSTM
line 14
model.add(LSTM(128,input_shape=(x_train.shape[1:]),activation='relu',return_sequences=True))
line 17
model.add(LSTM(128),activation='relu')
Digit_Recognision.ipnb
gives the detailed explanation of each steps along with a sample output.