Spaces:
Sleeping
Sleeping
Commit History
-
moved app.py to fix the runtime error
157160f -
Add scripts for supervised fine-tuning and GRPO training
- Created `make_sft_dataset.py` to generate a supervised fine-tuning dataset from expert plans. - Implemented `train_sft.py` for supervised fine-tuning of the model using the generated dataset. - Developed `train_grpo.py` to train the model using GRPO starting from the SFT checkpoint, including fresh trainable LoRA parameters.adcad94 -
round2: inference and planner critic design
786f4c0
-
Simplify main function by removing host and port arguments in app.py
d932af4 -
Refactor requirements file structure
401a2d6 -
Fixed architecture step function
Verified signatureabc93f5 -
Update server/app.py
Verified signaturereverted app server config27202c4 -
Update server/app.py
Verified signatureUpdated server configbd60fcd -
Fixed deployment bugs
Verified signature2c87f99 -
Delete server/Dockerfile
Verified signatureRemving docker5a8383f -
Upload 4 files
Verified signatureconfigured docker and env files3544a3b -
Create server/app.py
Verified signaturecreated app.py9c2e75a -
Delete server
Verified signature82ff700 -
Create server/
Verified signaturecb5582a