0

I'm using the following link, to start working on Google Cloud Vision platform. I tried the following steps:

1) Created the project

2) Created the bucket

3) I also ran the following code:

4) Ran each code given in the tutorial line-by-line

When I'm running this part of the code, it's throwing the following error.

gcloud ml-engine jobs submit training `whoami`_object_detection_`date +%s` \
--job-dir=gs://$BUCKET_NAME/data \
--packages dist/object_detection-0.1.tar.gz,slim/dist/slim-0.1.tar.gz \
--module-name object_detection.train \
--region us-central1 \
--config object_detection/samples/cloud/cloud.yml \
-- \
--train_dir=gs://$BUCKET_NAME/data \
--pipeline_config_path=${BUCKET_NAME}/data/faster_rcnn_resnet101_pets.config

Error:

ERROR: gcloud crashed (IOError): [Errno 2] No such file or directory: 'slim/dist/slim-0.1.tar.gz'

If you would like to report this issue, please run the following command: gcloud feedback

To check gcloud for common problems, please run the following command: gcloud info --run-diagnostics

I'm configuring the pipeline as following:

sed -i "s|gcloudstuff/visionex/models|"${BUCKET_NAME}"/data|g" object_detection/samples/configs/faster_rcnn_resnet101_pets.config
Nakilon
  • 34,866
  • 14
  • 107
  • 142
Beta
  • 1,638
  • 5
  • 33
  • 67
  • try to look into your TF setup - as the errors says no 'slim' directory, which you must have set during installation. check if you have the dir slim/dist/ ? – Nabeel Ahmed Jun 19 '17 at 08:02
  • did you run the TF object detection API installation test - python object_detection/builders/model_builder_test.py ? – Nabeel Ahmed Jun 19 '17 at 08:04
  • @NabeelAhmed: Thanks for your comment! I was following the tutorial almost exactly it's mentioned. Since there was no test mentioned I didn't try any test. Could you specify how to do the test. Since I downloaded the "model" folder, it has the slim directory. In the model execution code 2 packages are required "object detection" and "slim". I can't able to find how to do this part. – Beta Jun 19 '17 at 08:15
  • I copied the statement from in the installing the TensorFlow Object Detection API section. Verify if you have the slim/ added in PYTHONPATH - execute the command - python -c "import sys; print '\n'.join(sys.path)" - and share look in it – Nabeel Ahmed Jun 19 '17 at 08:37

0 Answers0