CODEWITHSUNDEEP
pythonkerastensorflow-lite
T.K
T.K

Reputation: 464

How to convert just a h5 file to a tflite file?

I'm trying to run license plate detection on Android. So first of all I find this tutorial: https://medium.com/@quangnhatnguyenle/detect-and-recognize-vehicles-license-plate-with-machine-learning-and-python-part-1-detection-795fda47e922 which is really great by the way.

In the tutorial, we can find wpod-net.h5 so I tried to convert it to TensorFlow lite using the following :

import tensorflow as tf

model = tf.keras.models.load_model('wpod-net.h5')
converter = tf.lite.TFLiteConverter.from_keras_model(model)
converter.post_training_quantize = True
tflite_model = converter.convert()
open("wpod-net.tflite", "wb").write(tflite_model)

But when I run this I have this error :

  File "converter.py", line 3, in <module>
    model = tf.keras.models.load_model('License_character_recognition.h5')
  File "/home/.local/lib/python3.8/site-packages/tensorflow/python/keras/saving/save.py", line 184, in load_model
    return hdf5_format.load_model_from_hdf5(filepath, custom_objects,
  File "/home/.local/lib/python3.8/site-packages/tensorflow/python/keras/saving/hdf5_format.py", line 175, in load_model_from_hdf5
    raise ValueError('No model found in config file.')
ValueError: No model found in config file.

I also tried using the API tflite_convert --keras_model_file=License_character_recognition.h5 --output_file=test.tflite but it gave me the same error.

Does that mean that if I didn't train the model myself I can't convert it to tflite ? Or is there another way to convert the .h5?

Upvotes: 1

Views: 3523

Answers (1)

Alex K.
Alex K.

Reputation: 861

TensorFlow Lite model incorporates both weights and model code itself. You need to load Keras model(with weights) and then you will be able to convert into tflite model.

Get a copy of authors' repo, and execute get-networks.sh. You need only data/lp-detector/wpod-net_update1.h5 for license plates detector so you can stop download earlier.

Dive a bit into code and you can find prepared load model function at keras utils.

After you get a model object, you can convert it into tflite.

Python3, TF2.4 tested:

import sys, os
import tensorflow as tf
import traceback

from os.path                    import splitext, basename

print(tf.__version__)

mod_path = "data/lp-detector/wpod-net_update1.h5"

def load_model(path,custom_objects={},verbose=0):
    #from tf.keras.models import model_from_json

    path = splitext(path)[0]
    with open('%s.json' % path,'r') as json_file:
        model_json = json_file.read()
    model = tf.keras.models.model_from_json(model_json, custom_objects=custom_objects)
    model.load_weights('%s.h5' % path)
    if verbose: print('Loaded from %s' % path)
    return model

keras_mod = load_model(mod_path)

converter = tf.lite.TFLiteConverter.from_keras_model(keras_mod)
tflite_model = converter.convert()

# Save the TF Lite model.
with tf.io.gfile.GFile('model.tflite', 'wb') as f:
    f.write(tflite_model)

Good luck!

Upvotes: 3

Related Questions