[pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

docs/source/_toctree.yml

@@ -10,3 +10,8 @@
   - local: getting_started_real_world_robot
     title: Getting Started with Real-World Robots
   title: "Tutorials"
+- sections:
+  - local: smolvla
+    title: Use SmolVLA
+  title: "Policies"
+

docs/source/smolvla.mdx

@@ -0,0 +1,91 @@
+# Use SmolVLA
+
+SmolVLA is designed to be easy to use and integrate—whether you're finetuning on your own data or plugging it into an existing robotics stack.
+
+<p align="center">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/640e21ef3c82bd463ee5a76d/aooU0a3DMtYmy_1IWMaIM.png" alt="SmolVLA architecture." width="500"/>
+  <br/>
+  <em>Figure 2. SmolVLA takes as input a sequence of RGB images from multiple cameras, the robot’s current sensorimotor state, and a natural language instruction. The VLM encodes these into contextual features, which condition the action expert to generate a continuous sequence of actions.</em>
+</p>
+
+###  Install
+
+First, install the required dependencies:
+
+```python
+git clone https://github.com/huggingface/lerobot.git
+cd lerobot
+pip install -e ".[smolvla]"
+```
+
+### Finetune the pretrained model
+Use [`smolvla_base`](https://hf.co/lerobot/smolvla_base), our pretrained 450M model, with the lerobot training framework:
+
+```python
+python lerobot/scripts/train.py \
+  --policy.path=lerobot/smolvla_base \
+  --dataset.repo_id=lerobot/svla_so100_stacking \
+  --batch_size=64 \
+  --steps=200000
+```
+
+<p align="center">
+  <img src="https://cdn-uploads.huggingface.co/production/uploads/640e21ef3c82bd463ee5a76d/S-3vvVCulChREwHDkquoc.gif" alt="Comparison of SmolVLA across task variations." width="500"/>
+  <br/>
+  <em>Figure 1: Comparison of SmolVLA across task variations. From left to right: (1) asynchronous pick-place cube counting, (2) synchronous pick-place cube counting, (3) pick-place cube counting under perturbations, and (4) generalization on pick-and-place of the lego block with real-world SO101.</em>
+</p>
+
+
+### Train from scratch
+
+​​If you'd like to build from the architecture (pretrained VLM + action expert) rather than a pretrained checkpoint:
+
+```python
+python lerobot/scripts/train.py \
+  --policy.type=smolvla \
+  --dataset.repo_id=lerobot/svla_so100_stacking \
+  --batch_size=64 \
+  --steps=200000
+```
+You can also load `SmolVLAPolicy` directly:
+
+```python
+from lerobot.common.policies.smolvla.modeling_smolvla import SmolVLAPolicy
+policy = SmolVLAPolicy.from_pretrained("lerobot/smolvla_base")
+```
+
+## Evaluate the pretrained policy and run it in real-time
+
+If you want to record the evaluation process and safe the videos on the hub, login to your HF account by running:
+
+```python
+huggingface-cli login --token ${HUGGINGFACE_TOKEN} --add-to-git-credential
+```
+
+Store your Hugging Face repository name in a variable to run these commands:
+
+```python
+HF_USER=$(huggingface-cli whoami | head -n 1)
+echo $HF_USER
+```
+Now, indicate the path to the policy, which is `lerobot/smolvla_base` in this case, and run:
+
+```python
+
+python lerobot/scripts/control_robot.py \
+  --robot.type=so100 \
+  --control.type=record \
+  --control.fps=30 \
+  --control.single_task="Grasp a lego block and put it in the bin." \
+  --control.repo_id=${HF_USER}/eval_svla_base_test \
+  --control.tags='["tutorial"]' \
+  --control.warmup_time_s=5 \
+  --control.episode_time_s=30 \
+  --control.reset_time_s=30 \
+  --control.num_episodes=10 \
+  --control.push_to_hub=true \
+  --control.policy.path=lerobot/smolvla_base
+
+```
+
+Depending on your evaluation setup, you can configure the duration and the number of episodes to record for your evaluation suite.

lerobot/__init__.py

@@ -168,7 +168,7 @@ available_datasets = sorted(
 )
 
 # lists all available policies from `lerobot/common/policies`
-available_policies = ["act", "diffusion", "tdmpc", "vqbet", "smolvla"]
+available_policies = ["act", "diffusion", "tdmpc", "vqbet"]
 
 # lists all available robots from `lerobot/common/robot_devices/robots`
 available_robots = [

lerobot/common/policies/smolvla/modeling_smolvla.py

@@ -662,7 +662,6 @@ class VLAFlowMatching(nn.Module):
             self.config.max_period,
             device=device,
         )
-
         time_emb = time_emb.type(dtype=dtype)
 
         time_emb = time_emb[:, None, :].expand_as(action_emb)

lerobot/common/robot_devices/control_utils.py

@@ -272,6 +272,7 @@ def control_loop(
                 action = {"action": action}
 
         if dataset is not None:
+            observation = {k: v for k, v in observation.items() if k not in ["task", "robot_type"]}
             frame = {**observation, **action, "task": single_task}
             dataset.add_frame(frame)
 

tests/test_available.py

@@ -21,7 +21,6 @@ import pytest
 import lerobot
 from lerobot.common.policies.act.modeling_act import ACTPolicy
 from lerobot.common.policies.diffusion.modeling_diffusion import DiffusionPolicy
-from lerobot.common.policies.smolvla.modeling_smolvla import SmolVLAPolicy
 from lerobot.common.policies.tdmpc.modeling_tdmpc import TDMPCPolicy
 from lerobot.common.policies.vqbet.modeling_vqbet import VQBeTPolicy
 from tests.utils import require_env
@@ -46,7 +45,7 @@ def test_available_policies():
     This test verifies that the class attribute `name` for all policies is
     consistent with those listed in `lerobot/__init__.py`.
     """
-    policy_classes = [ACTPolicy, DiffusionPolicy, TDMPCPolicy, VQBeTPolicy, SmolVLAPolicy]
+    policy_classes = [ACTPolicy, DiffusionPolicy, TDMPCPolicy, VQBeTPolicy]
     policies = [pol_cls.name for pol_cls in policy_classes]
     assert set(policies) == set(lerobot.available_policies), policies