from tensorflow.keras.layers import Input, Dense, LSTM, Bidirectional, Conv1D
from tensorflow.keras.layers import Flatten, Dropout
from tensorflow.keras.models import Model
from tensorflow.keras.optimizers import Adam
import numpy as np
from time import time

def timeit(func, iterations, *args):
    t0 = time()
    for _ in range(iterations):
        print('.')
        func(*args)
    print("%.4f sec" % (time() - t0))

batch_size = 32
batch_shape = (batch_size, 400, 16)
ipt   = Input(batch_shape=batch_shape)
x     = Bidirectional(LSTM(512, activation='relu', return_sequences=True))(ipt)
x     = LSTM(512, activation='relu', return_sequences=True)(ipt)
x     = Conv1D(128, 400, strides=1, padding='same')(x)
x     = Flatten()(x)
x     = Dense(256, activation='relu')(x)
x     = Dropout(0.5)(x)
x     = Dense(128, activation='relu')(x)
x     = Dense(64,  activation='relu')(x)
out   = Dense(1,  activation='sigmoid')(x)
model = Model(ipt, out)

X = np.random.randn(*batch_shape)
Y = np.random.randint(0, 2, (batch_size, 1))

model.compile(Adam(lr=1e-4), 'binary_crossentropy')

model.train_on_batch(X, Y)
timeit(model.train_on_batch, 4, X, Y)