admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-05-23 04:30:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

chore: apply prettier formatting to docs

Browse Source 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Pepijn
2026-04-08 12:48:35 +02:00
parent e35b485f4b
commit 28d353e0be
2 changed files with 32 additions and 31 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/env_processor.mdx
+1 -1
View File
@@ -145,7 +145,7 @@ class LiberoVelocityProcessorStep(ObservationProcessorStep):
state = torch.cat([eef_pos, eef_axisangle, eef_vel, state = torch.cat([eef_pos, eef_axisangle, eef_vel,
gripper_pos, gripper_vel], dim=-1) # 14D gripper_pos, gripper_vel], dim=-1) # 14D
return state return state
```` ```
### 4. **Cleaner Environment Code** ### 4. **Cleaner Environment Code**
docs/source/evaluation.mdx
+5 -4
View File
@@ -29,7 +29,7 @@ lerobot-eval \
## Key flags ## Key flags
| Flag | Default | Description | | Flag | Default | Description |
|---|---|---| | ----------------------- | -------------- | ------------------------------------------------------------------------------------- |
| `--policy.path` | required | Hub repo ID or local path to a pretrained model | | `--policy.path` | required | Hub repo ID or local path to a pretrained model |
| `--env.type` | required | Benchmark name (`pusht`, `libero`, `metaworld`, etc.) | | `--env.type` | required | Benchmark name (`pusht`, `libero`, `metaworld`, etc.) |
| `--env.task` | varies | Task or suite name (e.g. `libero_spatial`, `libero_10`) | | `--env.task` | varies | Task or suite name (e.g. `libero_spatial`, `libero_10`) |
@@ -60,7 +60,7 @@ See each benchmark's documentation ([LIBERO](libero), [Meta-World](metaworld)) f
`batch_size` controls how many environments run in parallel within a single `VectorEnv`: `batch_size` controls how many environments run in parallel within a single `VectorEnv`:
| `batch_size` | Behavior | | `batch_size` | Behavior |
|---|---| | ------------- | -------------------------------------------------------------------- |
| `0` (default) | Auto-tune: `floor(cpu_cores × 0.7)`, capped by `n_episodes` and `64` | | `0` (default) | Auto-tune: `floor(cpu_cores × 0.7)`, capped by `n_episodes` and `64` |
| `1` | Single environment, synchronous. Useful for debugging | | `1` | Single environment, synchronous. Useful for debugging |
| `N` | N environments step in parallel via `AsyncVectorEnv` | | `N` | N environments step in parallel via `AsyncVectorEnv` |
@@ -68,6 +68,7 @@ See each benchmark's documentation ([LIBERO](libero), [Meta-World](metaworld)) f
When `batch_size > 1` and `use_async_envs=true`, each environment runs in its own subprocess via Gymnasium's `AsyncVectorEnv`. This parallelizes the simulation stepping (the main bottleneck), while the policy runs a single batched forward pass on GPU. When `batch_size > 1` and `use_async_envs=true`, each environment runs in its own subprocess via Gymnasium's `AsyncVectorEnv`. This parallelizes the simulation stepping (the main bottleneck), while the policy runs a single batched forward pass on GPU.
**Example:** On a 16-core machine with `n_episodes=100`: **Example:** On a 16-core machine with `n_episodes=100`:
- Auto batch_size = `floor(16 × 0.7)` = `11` - Auto batch_size = `floor(16 × 0.7)` = `11`
- 11 environments step simultaneously → ~11× faster than sequential - 11 environments step simultaneously → ~11× faster than sequential
@@ -92,7 +93,7 @@ For multi-task benchmarks (e.g. LIBERO with 10 tasks), environments are wrapped
### Tuning for speed ### Tuning for speed
| Situation | Recommendation | | Situation | Recommendation |
|---|---| | ------------------------------ | ----------------------------------------------------- |
| Slow eval, low GPU utilization | Increase `batch_size` (or leave at auto) | | Slow eval, low GPU utilization | Increase `batch_size` (or leave at auto) |
| Out of memory (system RAM) | Decrease `batch_size` | | Out of memory (system RAM) | Decrease `batch_size` |
| Out of GPU memory | Decrease `batch_size`, or use `--policy.use_amp=true` | | Out of GPU memory | Decrease `batch_size`, or use `--policy.use_amp=true` |
@@ -108,7 +109,7 @@ Results are written to `output_dir` (default: `outputs/eval/<date>/<time>_<job_n
### Metrics ### Metrics
| Metric | Description | | Metric | Description |
|---|---| | ---------------- | -------------------------------------------------------------------- |
| `pc_success` | Success rate (%). Based on `info["is_success"]` from the environment | | `pc_success` | Success rate (%). Based on `info["is_success"]` from the environment |
| `avg_sum_reward` | Mean cumulative reward per episode | | `avg_sum_reward` | Mean cumulative reward per episode |
| `avg_max_reward` | Mean peak reward per episode | | `avg_max_reward` | Mean peak reward per episode |