OlafenwaMoses / ImageAI

A python library built to empower developers to build applications and systems with self-contained Computer Vision capabilities
https://www.genxr.co/#products
MIT License
8.58k stars 2.19k forks source link

When i convert h5 file to pb file, tensorflow api project get error #430

Open gokalpgithub opened 4 years ago

gokalpgithub commented 4 years ago

OlanfenwaMoses thanks for everything, you almost saved my life :) but i need help for last step. How can i convert my h5 file to pb. I want to use my model at tensorflow project. Is the same efficiency when the h5 file is converted to a pb file ?

When i integrated my pb model to tensorflow object detection api, i display this error

Traceback (most recent call last): File "inference_video_face.py", line 44, in tf.import_graph_def(od_graph_def, name='') File "C:\Users\gokalprogram\Miniconda3\lib\site-packages\tensorflow\python\util\deprecation.py", line 507, in new_func return func(*args, **kwargs) File "C:\Users\gokalprogram\Miniconda3\lib\site-packages\tensorflow\python\framework\importer.py", line 431, in import_graph_def raise ValueError(str(e)) ValueError: Input 0 of node bnorm_0/cond/ReadVariableOp/Switch was passed float from bnorm_0/gamma:0 incompatible with expected resource.

lumauf commented 4 years ago

Same, is it easy/possible to use a trained custom model (dataset = SVHN and model Yolov3 for multi digit recognition) with tensorflow, tensorflow.js, Tensorflow lite ?