aws / aws/amazon-sagemaker-examples

bandits_statlog_vw_customEnv: Customer Error: Unable to load model (caused by JSONDecodeError)

Open
#1,304 1 comment 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
11k
Forks
7k
Avg merge
8h 29m
Merged PRs (30d)
8

Description

The bandits statlog Vowpal wabbit example is not able to load the model at the step

bandits_experiment.deploy_model(model_id=bandits_experiment.last_trained_model_id)

Here is trace:

algo-1-yklna_1 | Caused by: Expecting value: line 1 column 1 (char 0)
algo-1-yklna_1 | Traceback (most recent call last):
algo-1-yklna_1 | File "/usr/local/lib/python3.6/dist-packages/vw_serving/serve.py", line 235, in get_model
algo-1-yklna_1 | quiet_mode=True)
algo-1-yklna_1 | File "/usr/local/lib/python3.6/dist-packages/vw_serving/vw_agent.py", line 289, in load_model
algo-1-yklna_1 | metadata = json.loads(metadata)
algo-1-yklna_1 | File "/usr/lib/python3.6/json/__init__.py", line 354, in loads
algo-1-yklna_1 | return _default_decoder.decode(s)
algo-1-yklna_1 | File "/usr/lib/python3.6/json/decoder.py", line 339, in decode
algo-1-yklna_1 | obj, end = self.raw_decode(s, idx=_w(s, 0).end())
algo-1-yklna_1 | File "/usr/lib/python3.6/json/decoder.py", line 357, in raw_decode
algo-1-yklna_1 | raise JSONDecodeError("Expecting value", s, err.value) from None
algo-1-yklna_1 | json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

Contributor guide

Open the contributing guide

Research direction

Start at the statlog Vowpal Wabbit example and the bandits_experiment.deploy_model(model_id=bandits_experiment.last_trained_model_id) call. Read the traceback through vw_serving.get_model and load_model, then reproduce the deployment; done means the example loads the trained model without the JSONDecodeError.

Written by the indexing model from the issue text.

Assessment

Tech stack
aws, jupyter-notebook, python
Domain
cloud, machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.