InvalidArgumentError: cannot compute MatMul as input #1(zero-based) was expected to be a double tensor but is a float tensor [Op:MatMul]

Question

Here is my code:

training_loss=[]
val_loss=[]

for epoch in range(1000):
  loss,v_loss = train(X_train1, y_train1,X_val1,y_val1)
  training_loss.append(loss)
  val_loss.append(v_loss)
  
  print('Epoch %d: Training Loss = %.4f, validation_loss= %.4f' % (epoch, float(loss),float(v_loss)))


print('Final Weights after 100 epochs:')
print('###############################################################################')
print(weights)

print('Final Bias after 100 epochs:')
print('###############################################################################')
print(bias)

I tried casting X and y into float32, like below. Y = Y.astype('float32') X_train = X_train.astype('float32')

But I still get the error.

I was expecting this to print to the final weights and bias.