Question

我要将body-pix移植到Python和C ++，并想将tensorflow.js的body-pix预先训练的模型导出到tensorflow冻结图中。有可能吗？

我已经下载了以下文件，并尝试使用tensorflowjs_converter进行转换，但是没有用。

结果在这里。

$ tensorflowjs_converter --input_format tfjs_layers_model --output_format keras posenet_mobilenet_025_partmap/model.json test.h5

Traceback (most recent call last):
  File "/home/xxx/anaconda3/envs/tfjs_test2/bin/tensorflowjs_converter", line 10, in <module>
    sys.exit(main())
  File "/home/xxx/anaconda3/envs/tfjs_test2/lib/python3.6/site-packages/tensorflowjs/converters/converter.py", line 368, in main
    FLAGS.output_path)
  File "/home/xxx/anaconda3/envs/tfjs_test2/lib/python3.6/site-packages/tensorflowjs/converters/converter.py", line 169, in dispatch_tensorflowjs_to_keras_h5_conversion
    model = keras_tfjs_loader.load_keras_model(config_json_path)
  File "/home/xxx/anaconda3/envs/tfjs_test2/lib/python3.6/site-packages/tensorflowjs/converters/keras_tfjs_loader.py", line 218, in load_keras_model
    use_unique_name_scope=use_unique_name_scope)
  File "/home/xxx/anaconda3/envs/tfjs_test2/lib/python3.6/site-packages/tensorflowjs/converters/keras_tfjs_loader.py", line 65, in _deserialize_keras_model
    model = keras.models.model_from_json(json.dumps(model_topology_json))
  File "/home/xxx/anaconda3/envs/tfjs_test2/lib/python3.6/site-packages/tensorflow/python/keras/saving/model_config.py", line 96, in model_from_json
    return deserialize(config, custom_objects=custom_objects)
  File "/home/xxx/anaconda3/envs/tfjs_test2/lib/python3.6/site-packages/tensorflow/python/keras/layers/serialization.py", line 81, in deserialize
    layer_class_name = config['class_name']
KeyError: 'class_name'

转换器版本在这里。

tensorflowjs 1.0.1

Dependency versions:
  keras 2.2.4-tf
  tensorflow 2.0.0-dev20190405

在Ubuntu 16.04 LTS和anaconda 3。我已经尝试过tensorflowjs 0.8.5，但是也没有用。

如果您告诉我如何进行转换，将很有帮助。 keras格式或tensorflow冻结图都可以。我认为两者可以相互转换。

Answer 1

下载model.json文件

例如：https://storage.googleapis.com/tfjs-models/savedmodel/bodypix/resnet50/float/model-stride16.json

从manifest.json下载相应的权重

https://storage.googleapis.com/tfjs-models/savedmodel/bodypix/resnet50/float/manifest.json

安装tfjs_graph_converter

来自https://github.com/ajaichemmanam/tfjs-to-tf

将模型转换为.pb文件

tfjs_graph_converter路径/ to / js / model路径/to/frozen/model.pb

这里是将POSENET转换为keras h5模型的示例。 https://github.com/tensorflow/tfjs/files/3943875/posenet.zip 用相同的方法可以使用bodypix模型并将其转换。

如何将tfjs的body-pix模型转换为keras h5或tensorflow冻结图

1 个答案: