update direct stepwise train, pass trainning with mock one step trace duplicated 52 times; reward OK, curve improved as the step grows.

This commit is contained in:
2025-06-30 15:06:38 +00:00
parent 5e632c53ac
commit 4295f30f9a
56 changed files with 1409 additions and 1 deletions

4
.gitignore vendored
View File

@@ -4,4 +4,6 @@ sample/sample_output/
sample/tmp_dataset_gen/
sample/output_data/
*__pycache__/
*__pycache__/
*checkpoint*