search

tensorflow / models

Models and examples built with TensorFlow
Other
76.95k stars 45.8k forks source link

rfcn_resnet101_coco model optimize issue #7576

Open jhxing opened 4 years ago

jhxing commented 4 years ago

Hi, I have succeeded in optimizing rfcn_resnet101_coco_2018_01_28(downloaded from model zoo) to IR when I use the following command:

/opt/intel/computer_vision_sdk/deployment_tools/model_optimizer/mo_tf.py --input_model frozen_inference_graph2018.pb --tensorflow_use_custom_operations_config /opt/intel/computer_vision_sdk/deployment_tools/model_optimizer/extensions/front/tf/rfcn_support.json --tensorflow_object_detection_api_pipeline_config pipeline.config --reverse_input_channels --data_type=FP32 --input_shape=[1,600,600,3]

While, I reuse the command to optimize rfcn_resnet101_coco_11_06_2017 (downloaded from Tensorflow Object Detection API) to IR, I got the this ERROR:

[ ERROR ] Exception occurred during running replacer "ObjectDetectionAPIDetectionOutputReplacement" (<class 'extensions.front.tf.ObjectDetectionAPI.ObjectDetectionAPIDetectionOutputReplacement'>): Found the following nodes '[]' with name 'proposals' but there should be exactly 1. Looks like ObjectDetectionAPIProposalReplacement replacement didn't work.

I tried diffetent rfcn_support_api.json files(v1.10, v1.13)， still got this ERROR. Anyone could help me solve this issue ?

tensorflowbutler commented 4 years ago

Thank you for your post. We noticed you have not filled out the following field in the issue template. Could you update them if they are relevant in your case, or leave them as N/A? Thanks. What is the top-level directory of the model you are using Have I written custom code OS Platform and Distribution TensorFlow installed from TensorFlow version Bazel version CUDA/cuDNN version GPU model and memory Exact command to reproduce

jhxing commented 4 years ago

Thank you for your post. We noticed you have not filled out the following field in the issue template. Could you update them if they are relevant in your case, or leave them as N/A? Thanks. What is the top-level directory of the model you are using Have I written custom code OS Platform and Distribution TensorFlow installed from TensorFlow version Bazel version CUDA/cuDNN version GPU model and memory Exact command to reproduce

Thank you for your attention. When I optimized the rfcn_resnet101_coco_11_06_2017 (downloaded from Tensorflow Object Detection API), the top-level directory of the model is:

/home/smc/hongxing/Tenserflow/models/research/object_detection/rfcn_resnet101_coco_11_06_2017 the model was downloaded from:http://download.tensorflow.org/models/object_detection/rfcn_resnet101_coco_11_06_2017.tar.gz and the tensorflow version is v1.14=gpu , and the Object Detection API was doenloaded from: https://github.com/tensorflow/models/tree/master/research