model_h5_to_tflite9.py

def evaluate_model(interpreter):
  input_index = interpreter.get_input_details()[0]["index"]
  output_index = interpreter.get_output_details()[0]["index"]

  # Run predictions on ever y image in the "test" dataset.
  prediction_digits = []
  for i, photo_id in enumerate(df_photo_ids['test'].photo_id):
    if i % 1000 == 0:
      print('Evaluated on {n} results so far.'.format(n=i))
    # Pre-processing: add batch dimension and convert to float32 to match with
    # the model's input data format.
    test_image = fetch_img_by_photo_id_and_resize(photo_id, (params["img_size"], params["img_size"]))
    test_image = np.array([test_image], dtype=np.float32)
    interpreter.set_tensor(input_index, test_image)

    # Run inference.
    interpreter.invoke()

    # Post-processing: remove batch dimension and find the digit with highest probability.
    output = interpreter.tensor(output_index)
    digit = np.argmax(output()[0])
    print(i, photo_id, digit)
    prediction_digits.append(digit)

  print('\n')
  # Compare prediction results with ground truth labels to calculate accuracy.
  prediction_digits = np.array(prediction_digits)
  accuracy = (prediction_digits == df_photo_ids['test'].y).mean()
  return accuracy

interpreter = tf.lite.Interpreter(model_content=quantized_and_pruned_tflite_model)
interpreter.allocate_tensors()

test_accuracy = evaluate_model(interpreter)

print('Pruned and quantized TFLite test_accuracy:', test_accuracy)

with open(model_filename + '.tflite', 'wb') as f:
  f.write(quantized_and_pruned_tflite_model)