如何將.model 轉換為.tflite？

Question

我目前正在從事口罩檢測項目。 我不太了解機器學習，但這是我的問題陳述。 我必須在 Raspberry Pi 上運行我的項目，所以我決定將我的 model 轉換為 tflite 以提高 FPS。 我在這行'model.save（“mask_detector.model”，save_format =“h5”）'中訓練並創建了我的model，我使用了'res10_300x300_ssd_iter_140000.caffeemodel'。 但是，當我運行這些代碼行時

import tensorflow as tf
import numpy as np
from tensorflow import keras
from tensorflow.contrib import lite

converter = lite.TocoConverter.from_keras_model_file("D:\Face-Mask-Detection-master\mask_detector.model")
tflite_model = converter.convert()
open("linear.tflite", "wb").write(tflite_model)

我收到一個錯誤

ImportError: cannot import name ‘lite’. 

當我運行這些代碼行時

import tensorflow as tf
converter = tf.lite.TFLiteConverter.from_keras_model_file('mask_detector.model') 
tfmodel = converter.convert() 
open ("model.tflite" , "wb") .write(tfmodel)

我收到一個錯誤

File "converty.py", line 14, in <module>
    converter = tf.lite.TFLiteConverter.from_keras_model_file('mask_detector.model') 
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/lite/python/lite.py", line 747, in from_keras_model_file
    keras_model = _keras.models.load_model(model_file, custom_objects)
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/python/keras/saving/save.py", line 146, in load_model
    return hdf5_format.load_model_from_hdf5(filepath, custom_objects, compile)
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/python/keras/saving/hdf5_format.py", line 212, in load_model_from_hdf5
    custom_objects=custom_objects)
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/python/keras/saving/model_config.py", line 55, in model_from_config
    return deserialize(config, custom_objects=custom_objects)
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/python/keras/layers/serialization.py", line 89, in deserialize
    printable_module_name='layer')
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/python/keras/utils/generic_utils.py", line 181, in deserialize_keras_object
    config, module_objects, custom_objects, printable_module_name)
  File "/home/pi/.local/lib/python2.7/site-packages/tensorflow/python/keras/utils/generic_utils.py", line 166, in class_and_config_for_serialized_keras_object
    raise ValueError('Unknown ' + printable_module_name + ': ' + class_name)
ValueError: Unknown layer: Functional

如何將.model 轉換為.tflite？ 謝謝

Answer 1

TFLite 目前僅支持 TensorFlow 型號。 （這個問題可能是由於使用了res10_300x300_ssd_iter_140000.caffeemodel ）。 有沒有辦法先在 TensorFlow 中獲得 SavedModel 或 Keras model ？