Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Mspiz
New Contributor I
166 Views

Tensorflow mobile net_v2_coco issue OpenVino R3

Hi,

I have some issues on TensorFlow mobile net_v2_coco_2018_03_29 model, from downloader.py script.

python3 mo_tf.py --input_model ../model_downloader/object_detection/common/ssd_mobilenet_v2_coco/tf/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb --input_checkpoint ../model_downloader/object_detection/common/ssd_mobilenet_v2_coco/tf/ssd_mobilenet_v2_coco_2018_03_29/checkpoint  --model_name res_mobilenet-coco-v2  --output_dir ../my_ir_model/
Model Optimizer arguments:
Common parameters:
    - Path to the Input Model:     /opt/intel/computer_vision_sdk_2018.3.343/deployment_tools/model_optimizer/../model_downloader/object_detection/common/ssd_mobilenet_v2_coco/tf/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb
    - Path for generated IR:     /opt/intel/computer_vision_sdk_2018.3.343/deployment_tools/model_optimizer/../my_ir_model/
    - IR output name:     res_mobilenet-coco-v2
    - Log level:     ERROR
    - Batch:     Not specified, inherited from the model
    - Input layers:     Not specified, inherited from the model
    - Output layers:     Not specified, inherited from the model
    - Input shapes:     Not specified, inherited from the model
    - Mean values:     Not specified
    - Scale values:     Not specified
    - Scale factor:     Not specified
    - Precision of IR:     FP32
    - Enable fusing:     True
    - Enable grouped convolutions fusing:     True
    - Move mean values to preprocess section:     False
    - Reverse input channels:     False
TensorFlow specific parameters:
    - Input model in text protobuf format:     False
    - Offload unsupported operations:     False
    - Path to model dump for TensorBoard:     None
    - Update the configuration file with input/output node names:     None
    - Use configuration file used to generate the model with Object Detection API:     None
    - Operations to offload:     None
    - Patterns to offload:     None
    - Use the config file:     None
Model Optimizer version:     1.2.185.5335e231
[ ERROR ]  Cannot load input model: Unable to open table file /opt/intel/computer_vision_sdk_2018.3.343/deployment_tools/model_optimizer/../model_downloader/object_detection/common/ssd_mobilenet_v2_coco/tf/ssd_mobilenet_v2_coco_2018_03_29/checkpoint: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator?

 

 

Also if a give one parameter  frozen_graph.pb or I use --input_meta_graph it Fail.

However with my own model trained with TensorFlow  mo_tf.py work well.

0 Kudos
2 Replies
Monique_J_Intel
Employee
166 Views

Hi Carmine,

I've reproduced your error and just for your information when converting Tensoflow models with model optimizer you should supply either the frozen model in .pb or.pbtxt format or supply the metagraph(.ckpt-meta*) and checkpoint file(.ckpt) to model optimizer. Please use the following to convert your model:

sudo python3 mo.py --input_model ../model_downloader/object_detection/c
ommon/ssd_mobilenet_v2_coco/tf/ssd_mobilenet_v2_coco_2018_03_29/frozen_inference_graph.pb --tensorflow_use_custom_operations_config extensions/front/tf/ssd_v2_support.json --output="detection_boxes,detection_scores,num_detections" --tensorflow_object_detection_api_pipeline_config ../model_downloader/object_detection/common/ssd_mobilenet_v2_coco/tf/ssd_mobilenet_v2_coco_2018_03_29/pipeline.config

Kind Regards,

Monique Jones

Mspiz
New Contributor I
166 Views

Hi Monique,

Just tried it and it's work!

Can you clarify me:

Why in this case I need to use mo.py and not mo_tf.py?

Why I need of  ssd_v2_support.json and pipeline.config for transform this graph to IR?

And, which version of checkpoint in Tensorflow is supported by OpenVino R3? (Similar issue: .ckp: Data loss: not an sstable (bad magic number): perhaps your file is in a different file format and you need to use a different restore operator? on my own graph, could be the checkpoint version ?)

Thanks in advanced ! ;) 

 

Reply