Styjun

Strange LED behavior: Why is there a voltage over the LED with only one wire connected to it?

German phrase for 'suited and booted'

What kind of anatomy does a centaur have?

Killing a star safely

Has Peter Parker ever eaten bugs?

Why can't a country print its own money to spend it only abroad?

Why did modems have speakers?

Why is DC so, so, so Democratic?

Xcode 10.3 Installation

Why does the salt in the oceans not sink to the bottom?

How can I show that the speed of light in vacuum is the same in all reference frames?

Do you have to hide in order for other creatures to not know your location inside a fog cloud?

What is "It is x o'clock" in Japanese with subject

Are gangsters hired to attack people at a train station classified as a terrorist attack?

Can I make an Opportunity Attack during Time Stop?

Is there a way to shorten this while condition?

Short story where a flexible reality hardens to an unchanging one

Arduino Bluetooth communication is slowing down because of the 'for' loop!

Impact of throwing away fruit waste on a peak > 3200 m above a glacier

Why is the UH-60 tail rotor canted?

How can I calculate the cost of Skyss bus tickets

Can we have too many dialogue tags and follow up actions?

Giving blur shadow to plot

Is it better to merge "often" or only after completion do a big merge of feature branches?

How to deploy a model trained with a custom package on Google ML Engine?

Manage scikit-learn model in Google Cloud PlatformCan't train my Tensorflow detector model in Google CloudGcloud ml-engine: FAILED_PRECONDITION: Field: package_uris ErrorGoogle Machine Learning Engine Python 3Get the path of saved_model.pb after training on ML engineError creating model version using “gcloud ml-engine versions create”Google Cloud ML returns empty predictions with object detection modelTPOT model in google ml-engineGoogle Cloud Platform, Ml Engine, No module named sklearn_crfsuite.estimatorWhy does google ml-engine online predict return a different prediction from local predict?

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;

1

I'm able to train a scikit-learn model that includes custom functions on Google ML Engine with a call like this:

gcloud ml-engine jobs submit training "module_name" 
 --job-dir gs://my_bucket/scikit_learn_job_dir 
 --package-path ./my_package/ 
 --module-name my_package.trainer.train 
 --region us-east1 
 --runtime-version='1.12' 
 --python-version='3.5' 
 --scale-tier BASIC

This code trains the model and uploads a pickle file, model.pkl to gs://my_bucket/models.

However, I get an error when trying to deploy with :

gcloud ml-engine versions create "version_name" 
 --model model_name 
 --origin gs://my_bucket/models 
 --runtime-version='1.12' 
 --framework "SCIKIT_LEARN" 
 --python-version='3.5'

The error says that my_package cannot be loaded. Indeed, I can't find any option equivalent to --package-path in the version create command documentation.

python scikit-learn google-cloud-ml

share|improve this question

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  So the job is submitted/queued in ML Engine right? It would be super helpful if you could post your error logs from jobs in ML Engine.
  
  – Pramesh Bajracharya
  Mar 26 at 13:53
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Hello! The job is submitted in ML Engine and works fine. The model.pkl file is created and uploaded to GCS. The problem is when I create a model version from this .pkl file. The version has a red dot instead of a blue check mark and in the version details, the error is "Create Version failed. Bad model detected with error: "Failed to load model: Could not load the model: /tmp/model/0001/model.pkl. No module named 'my_package'. (Error code: 0)". This was expected, because I didn't send my_package with the version create call, but I didn't find how to!
  
  – Ajr
  Apr 2 at 2:12
  

  
  
  
  

add a comment | 

1

I'm able to train a scikit-learn model that includes custom functions on Google ML Engine with a call like this:

gcloud ml-engine jobs submit training "module_name" 
 --job-dir gs://my_bucket/scikit_learn_job_dir 
 --package-path ./my_package/ 
 --module-name my_package.trainer.train 
 --region us-east1 
 --runtime-version='1.12' 
 --python-version='3.5' 
 --scale-tier BASIC

This code trains the model and uploads a pickle file, model.pkl to gs://my_bucket/models.

However, I get an error when trying to deploy with :

gcloud ml-engine versions create "version_name" 
 --model model_name 
 --origin gs://my_bucket/models 
 --runtime-version='1.12' 
 --framework "SCIKIT_LEARN" 
 --python-version='3.5'

The error says that my_package cannot be loaded. Indeed, I can't find any option equivalent to --package-path in the version create command documentation.

python scikit-learn google-cloud-ml

share|improve this question

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges

• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  So the job is submitted/queued in ML Engine right? It would be super helpful if you could post your error logs from jobs in ML Engine.
  
  – Pramesh Bajracharya
  Mar 26 at 13:53
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Hello! The job is submitted in ML Engine and works fine. The model.pkl file is created and uploaded to GCS. The problem is when I create a model version from this .pkl file. The version has a red dot instead of a blue check mark and in the version details, the error is "Create Version failed. Bad model detected with error: "Failed to load model: Could not load the model: /tmp/model/0001/model.pkl. No module named 'my_package'. (Error code: 0)". This was expected, because I didn't send my_package with the version create call, but I didn't find how to!
  
  – Ajr
  Apr 2 at 2:12
  

  
  
  
  

add a comment | 

I'm able to train a scikit-learn model that includes custom functions on Google ML Engine with a call like this:

gcloud ml-engine jobs submit training "module_name" 
 --job-dir gs://my_bucket/scikit_learn_job_dir 
 --package-path ./my_package/ 
 --module-name my_package.trainer.train 
 --region us-east1 
 --runtime-version='1.12' 
 --python-version='3.5' 
 --scale-tier BASIC

This code trains the model and uploads a pickle file, model.pkl to gs://my_bucket/models.

However, I get an error when trying to deploy with :

gcloud ml-engine versions create "version_name" 
 --model model_name 
 --origin gs://my_bucket/models 
 --runtime-version='1.12' 
 --framework "SCIKIT_LEARN" 
 --python-version='3.5'

The error says that my_package cannot be loaded. Indeed, I can't find any option equivalent to --package-path in the version create command documentation.

python scikit-learn google-cloud-ml

share|improve this question

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges

gcloud ml-engine jobs submit training "module_name" 
 --job-dir gs://my_bucket/scikit_learn_job_dir 
 --package-path ./my_package/ 
 --module-name my_package.trainer.train 
 --region us-east1 
 --runtime-version='1.12' 
 --python-version='3.5' 
 --scale-tier BASIC

gcloud ml-engine versions create "version_name" 
 --model model_name 
 --origin gs://my_bucket/models 
 --runtime-version='1.12' 
 --framework "SCIKIT_LEARN" 
 --python-version='3.5'

share|improve this question

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges

share|improve this question

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges

asked Mar 26 at 13:42

AjrAjr

16155 bronze badges