vwxyzjn/cleanrl

run and experiment_name from docs/advanced/resume-training are undefined

Closed this issue · 5 comments

Melanol commented

Problem Description

Here. Also, am I supposed to use cleanrl_utils/resume.py instead of what is in the docs?

Checklist

vwxyzjn commented

The resume.py is something slightly more advanced (and kind of deprecated). You should rely on the pointers in docs.

Melanol commented

@vwxyzjn Docs are not helpful, as run and experiment_name are not defined anywhere before the following part, including algos:

if args.track and wandb.run.resumed:
    starting_update = run.summary.get("charts/update") + 1
    global_step = starting_update * args.batch_size
    api = wandb.Api()
    run = api.run(f"{run.entity}/{run.project}/{run.id}")
    model = run.file("agent.pt")
    model.download(f"models/{experiment_name}/")
    agent.load_state_dict(torch.load(
        f"models/{experiment_name}/agent.pt", map_location=device))
    agent.eval()
    print(f"resumed at update {starting_update}")
vwxyzjn commented

experiment_name is now run_name.

api = wandb.Api()
run = api.run(f"{run.entity}/{run.project}/{run.id}")
Melanol commented

run = api.run(f"{run.entity}/{run.project}/{run.id}")

You are trying to define run with run that is not defined.