0

按照https://www.tensorflow.org/tutorials/images/hub_with_keras上的教程生成一个文件model.h5。使用命令转换为 tensorflow-js

tensorflowjs_converter --input_format keras ./model.h5 /tmp/jsmodel/

失败了

例外：错误转储权重，重复权重名称变量

为什么会这样，如何解决？

MCVE

from __future__ import absolute_import, division, print_function
import tensorflow as tf
import tensorflow_hub as hub
from tensorflow.keras import layers
import numpy as np

data_root = tf.keras.utils.get_file(
'flower_photos','https://storage.googleapis.com/download.tensorflow.org/example_images/flower_photos.tgz',
   untar=True)

image_generator = tf.keras.preprocessing.image.ImageDataGenerator(rescale=1/255)
IMAGE_SHAPE = (224, 224)
image_data = image_generator.flow_from_directory(str(data_root), target_size=IMAGE_SHAPE)

feature_extractor_url = "https://tfhub.dev/google/tf2-preview/mobilenet_v2/feature_vector/2" #@param {type:"string"}
feature_extractor_layer = hub.KerasLayer(feature_extractor_url,
                                         input_shape=(224,224,3))

for image_batch, label_batch in image_data:
  print("Image batch shape: ", image_batch.shape)
  print("Labe batch shape: ", label_batch.shape)
  break

feature_extractor_layer.trainable = False

model = tf.keras.Sequential([
  feature_extractor_layer,
  layers.Dense(image_data.num_classes, activation='softmax')
])

model.compile(
  optimizer=tf.keras.optimizers.Adam(), 
  loss='categorical_crossentropy',
  metrics=['acc'])

steps_per_epoch = np.ceil(image_data.samples/image_data.batch_size)

history = model.fit(image_data, epochs=2, 
                    steps_per_epoch=steps_per_epoch)  # removed callback

model.save("/tmp/so_model.h5")

这失败了

RuntimeError：无法创建链接（名称已存在）

但模型已创建。上述调用tensorflowjs_converter --input_format keras /tmp/model.h5 /tmp/jsmodel失败

例外：错误转储权重，重复权重名称变量

更新：另请参阅使用 MobileNet 重新训练图像检测

4

0 回答 0