Snippet of RNN - GRU using Tensorflow

gru_example.py

import numpy as np
import tensorflow as tf
from tensorflow import keras
from tensorflow.keras import layers

# require for installation: !pip install -q git+https://github.com/tensorflow/docs
import tensorflow_docs as tfdocs
import tensorflow_docs.plots
import tensorflow_docs.modeling

gru_model = keras.Sequential([
    layers.GRU(32, return_sequences=True, input_shape=(n_steps, 1), activation='tanh'),
    layers.GRU(32, return_sequences=True, activation='tanh'),
    layers.Dropout(0.2),
    layers.GRU(32, return_sequences=True, activation='tanh'),
    layers.GRU(32, return_sequences=False, activation='tanh'),
    layers.Dropout(0.2),
    layers.Dense(1)
])

gru_model.compile(optimizer=tf.keras.optimizers.Adam(),
                  loss=tf.losses.MeanSquaredError(), metrics=['mae', 'mse'])

early_stopping = tf.keras.callbacks.EarlyStopping(monitor='val_loss',
                                                  patience=50,
                                                  mode='min')
gru_hist = gru_model.fit(
    X_train, y_train, epochs=500, validation_split=0.2,
    batch_size=32, verbose=0,
    callbacks=[tfdocs.modeling.EpochDots(), early_stopping]
)