CODEWITHSUNDEEP
pythontensorflowdeep-learningobject-detection
Glen Veigas
Glen Veigas

Reputation: 194

Python Failed to create a NewWriteableFile (tensorflow.python.framework.errors_impl.NotFoundError: Failed to create a NewWriteableFile: )

I am trying out the TensorFlow 2.0 Object Detection API, while I am trying to export the inference graph I get an error stating tensorflow.python.framework.errors_impl.NotFoundError: Failed to create a NewWriteableFile: .\exported-models\my_mobilenet_model\checkpoint\ckpt-0_temp_8cca890901704d9b9c1a0c5e959235fc/part-00000-of-00001.data-00000-of-00001.tempstate144711528486123721 : The system cannot find the path specified. ; No such process [Op:SaveV2]

I run the command: python .\exporter_main_v2.py --input_type image_tensor --pipeline_config_path .\models\my_ssd_mobilenet_v2_fpnlite\pipeline.config --trained_checkpoint_dir .\models\my_ssd_mobilenet_v2_fpnlite\ --output_directory .\exported-models\my_mobilenet_model

I do have a folder name exported-models in the directory I am running the command

Upvotes: 3

Views: 2066

Answers (1)

wader
wader

Reputation: 76

There is a limitation on the whole path length (<260). Pls. check your path to save the model, and you can manually try if you can create a folder as shown above, if not, pls. try to change the output_directory to another short path.

Upvotes: 5

Related Questions