Clarify SFT dataset in model card
Browse files
README.md
CHANGED
|
@@ -12,7 +12,13 @@ tags:
|
|
| 12 |
|
| 13 |
# OpenPI pi0.5 LoRA for sim2real mug into bowl
|
| 14 |
|
| 15 |
-
This repository stores the final OpenPI checkpoint for the `pi05_clutter_libero_lora` SFT run
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 16 |
|
| 17 |
## Checkpoint
|
| 18 |
|
|
@@ -39,7 +45,7 @@ checkpoints/pi05_clutter_libero_lora/sim2real_mug_from_sim_table_25000_30k/30000
|
|
| 39 |
|
| 40 |
- Config: `pi05_clutter_libero_lora`
|
| 41 |
- Experiment: `sim2real_mug_from_sim_table_25000_30k`
|
| 42 |
-
- Dataset: `IDEAS-Lab-Northwestern/sim2real-mug-into-bowl`
|
| 43 |
- Resume source: local `sim_table_lora/25000` checkpoint
|
| 44 |
- Final step: `30000`
|
| 45 |
- Batch size: `4`
|
|
|
|
| 12 |
|
| 13 |
# OpenPI pi0.5 LoRA for sim2real mug into bowl
|
| 14 |
|
| 15 |
+
This repository stores the final OpenPI checkpoint for the `pi05_clutter_libero_lora` SFT run.
|
| 16 |
+
|
| 17 |
+
## Dataset
|
| 18 |
+
|
| 19 |
+
This model was SFT-trained using the `IDEAS-Lab-Northwestern/sim2real-mug-into-bowl` dataset:
|
| 20 |
+
|
| 21 |
+
https://huggingface.co/datasets/IDEAS-Lab-Northwestern/sim2real-mug-into-bowl
|
| 22 |
|
| 23 |
## Checkpoint
|
| 24 |
|
|
|
|
| 45 |
|
| 46 |
- Config: `pi05_clutter_libero_lora`
|
| 47 |
- Experiment: `sim2real_mug_from_sim_table_25000_30k`
|
| 48 |
+
- Dataset: [`IDEAS-Lab-Northwestern/sim2real-mug-into-bowl`](https://huggingface.co/datasets/IDEAS-Lab-Northwestern/sim2real-mug-into-bowl)
|
| 49 |
- Resume source: local `sim_table_lora/25000` checkpoint
|
| 50 |
- Final step: `30000`
|
| 51 |
- Batch size: `4`
|