5 Commits

Author	SHA1	Message	Date
Simon Alibert	5877eb38ab	Enable backward compatibility testing on tdmpc	2024-05-03 10:04:49 +02:00
Simon Alibert	3a918b980f	WIP	2024-05-03 10:04:48 +02:00
Simon Alibert	7002813bbb	Cleanup, add output_dict	2024-05-03 10:04:48 +02:00
Simon Alibert	2726d1a6c9	Add get_policy_stats	2024-05-03 10:04:48 +02:00
Simon Alibert	6e15e9664b	WIP Add save_policy_to_safetensors	2024-05-03 10:04:48 +02:00