aws / aws/amazon-sagemaker-examples

'main_level/agent/main/online/network_1/ppo_head_0/policy' refers to an Operation not in the graph.

Open
#2,630 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
11k
Forks
7k
Avg merge
8h 29m
Merged PRs (30d)
8

Description

When running the amazon-sagemaker-examples/reinforcement_learning/rl_stock_trading_coach_customEnv/ does not save model and gives the following error:

KeyError: "The name 'main_level/agent/main/online/network_1/ppo_head_0/policy' refers to an Operation not in the graph."

This example uses dueling_q and not ppo. I believe something the _tf_save_model in coach_launcher.py has to be appropiately changed.

Contributor guide

Open the contributing guide

Research direction

Reproduce the failure in reinforcement_learning/rl_stock_trading_coach_customEnv/, then inspect coach_launcher.py and its _tf_save_model entry point. Trace how the dueling_q model is saved and verify the completed change saves the model without looking up the PPO operation that is absent from the graph.

Written by the indexing model from the issue text.

Assessment

Tech stack
aws, jupyter-notebook, python
Domain
machine-learning
Issue type
Bug
Difficulty
3/5
Estimated time
1-2 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.