[skip ci] chore(pre-commit): bump python version from 3.10 to python3.12

.github/workflows/pr_style_bot.yml

@@ -5,17 +5,50 @@ on:
   issue_comment:
     types: [created]
 
-permissions:
-  contents: write
-  pull-requests: write
+permissions: {}
+
+env:
+  PYTHON_VERSION: "3.10"
 
 jobs:
-  run-style-bot:
+  check-permissions:
     if: >
       contains(github.event.comment.body, '@bot /style') &&
       github.event.issue.pull_request != null
     runs-on: ubuntu-latest
+    outputs:
+      is_authorized: ${{ steps.check_user_permission.outputs.has_permission }}
+    steps:
+      - name: Check user permission
+        id: check_user_permission
+        uses: actions/github-script@v6
+        with:
+          script: |
+            const comment_user = context.payload.comment.user.login;
+            const { data: permission } = await github.rest.repos.getCollaboratorPermissionLevel({
+              owner: context.repo.owner,
+              repo: context.repo.repo,
+              username: comment_user
+            });
 
+            const authorized =
+              permission.permission === 'admin' ||
+              permission.permission === 'write';
+
+            console.log(
+              `User ${comment_user} has permission level: ${permission.permission}, ` +
+              `authorized: ${authorized} (admins & maintainers allowed)`
+            );
+
+            core.setOutput('has_permission', authorized);
+
+  run-style-bot:
+    needs: check-permissions
+    if: needs.check-permissions.outputs.is_authorized == 'true'
+    runs-on: ubuntu-latest
+    permissions:
+      contents: write
+      pull-requests: write
     steps:
       - name: Extract PR details
         id: pr_info
@@ -61,6 +94,8 @@ jobs:
 
       - name: Set up Python
         uses: actions/setup-python@v4
+        with:
+          python-version: ${{ env.PYTHON_VERSION }}
 
       - name: Get Ruff Version from pre-commit-config.yaml
         id: get-ruff-version
@@ -91,6 +126,7 @@ jobs:
           # Configure git with the Actions bot user
           git config user.name "github-actions[bot]"
           git config user.email "github-actions[bot]@users.noreply.github.com"
+          git config --local lfs.https://github.com/.locksverify false
 
           # Make sure your 'origin' remote is set to the contributor's fork
           git remote set-url origin "https://x-access-token:${GITHUB_TOKEN}@github.com/${HEADREPOFULLNAME}.git"

.pre-commit-config.yaml

@@ -1,7 +1,8 @@
 exclude: ^(tests/data)
 default_language_version:
-    python: python3.10
+    python: python3.12
 repos:
+  ##### Style / Misc. #####
   - repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v5.0.0
     hooks:
@@ -14,7 +15,7 @@ repos:
       - id: end-of-file-fixer
       - id: trailing-whitespace
   - repo: https://github.com/crate-ci/typos
-    rev: v1.29.10
+    rev: v1.30.0
     hooks:
       - id: typos
         args: [--force-exclude]
@@ -23,16 +24,24 @@ repos:
     hooks:
     -   id: pyupgrade
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.9.6
+    rev: v0.9.9
     hooks:
       - id: ruff
         args: [--fix]
       - id: ruff-format
+
+  ##### Security #####
   - repo: https://github.com/gitleaks/gitleaks
-    rev: v8.23.3
+    rev: v8.24.0
     hooks:
       - id: gitleaks
   - repo: https://github.com/woodruffw/zizmor-pre-commit
-    rev: v1.3.1
+    rev: v1.4.1
     hooks:
       - id: zizmor
+  - repo: https://github.com/PyCQA/bandit
+    rev: 1.8.3
+    hooks:
+    - id: bandit
+      args: ["-c", "pyproject.toml"]
+      additional_dependencies: ["bandit[toml]"]

examples/3_train_policy.py

@@ -85,7 +85,7 @@ def main():
     done = False
     while not done:
         for batch in dataloader:
-            batch = {k: v.to(device, non_blocking=True) for k, v in batch.items()}
+            batch = {k: (v.to(device) if isinstance(v, torch.Tensor) else v) for k, v in batch.items()}
             loss, _ = policy.forward(batch)
             loss.backward()
             optimizer.step()

lerobot/common/datasets/lerobot_dataset.py

@@ -13,6 +13,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import contextlib
 import logging
 import shutil
 from pathlib import Path
@@ -27,6 +28,7 @@ import torch.utils
 from datasets import concatenate_datasets, load_dataset
 from huggingface_hub import HfApi, snapshot_download
 from huggingface_hub.constants import REPOCARD_NAME
+from huggingface_hub.errors import RevisionNotFoundError
 
 from lerobot.common.constants import HF_LEROBOT_HOME
 from lerobot.common.datasets.compute_stats import aggregate_stats, compute_episode_stats
@@ -517,6 +519,7 @@ class LeRobotDataset(torch.utils.data.Dataset):
         branch: str | None = None,
         tags: list | None = None,
         license: str | None = "apache-2.0",
+        tag_version: bool = True,
         push_videos: bool = True,
         private: bool = False,
         allow_patterns: list[str] | str | None = None,
@@ -562,6 +565,11 @@ class LeRobotDataset(torch.utils.data.Dataset):
             )
             card.push_to_hub(repo_id=self.repo_id, repo_type="dataset", revision=branch)
 
+        if tag_version:
+            with contextlib.suppress(RevisionNotFoundError):
+                hub_api.delete_tag(self.repo_id, tag=CODEBASE_VERSION, repo_type="dataset")
+            hub_api.create_tag(self.repo_id, tag=CODEBASE_VERSION, revision=branch, repo_type="dataset")
+
     def pull_from_repo(
         self,
         allow_patterns: list[str] | str | None = None,

lerobot/common/datasets/utils.py

@@ -31,6 +31,7 @@ import packaging.version
 import torch
 from datasets.table import embed_table_storage
 from huggingface_hub import DatasetCard, DatasetCardData, HfApi
+from huggingface_hub.errors import RevisionNotFoundError
 from PIL import Image as PILImage
 from torchvision import transforms
 
@@ -325,6 +326,19 @@ def get_safe_version(repo_id: str, version: str | packaging.version.Version) ->
     )
     hub_versions = get_repo_versions(repo_id)
 
+    if not hub_versions:
+        raise RevisionNotFoundError(
+            f"""Your dataset must be tagged with a codebase version.
+            Assuming _version_ is the codebase_version value in the info.json, you can run this:
+            ```python
+            from huggingface_hub import HfApi
+
+            hub_api = HfApi()
+            hub_api.create_tag("{repo_id}", tag="_version_", repo_type="dataset")
+            ```
+            """
+        )
+
     if target_version in hub_versions:
         return f"v{target_version}"
 

lerobot/common/datasets/v21/convert_dataset_v20_to_v21.py

@@ -57,7 +57,7 @@ def convert_dataset(
     dataset.meta.info["codebase_version"] = CODEBASE_VERSION
     write_info(dataset.meta.info, dataset.root)
 
-    dataset.push_to_hub(branch=branch, allow_patterns="meta/")
+    dataset.push_to_hub(branch=branch, tag_version=False, allow_patterns="meta/")
 
     # delete old stats.json file
     if (dataset.root / STATS_PATH).is_file:

lerobot/common/utils/utils.py

@@ -17,6 +17,7 @@ import logging
 import os
 import os.path as osp
 import platform
+import subprocess
 from copy import copy
 from datetime import datetime, timezone
 from pathlib import Path
@@ -165,23 +166,31 @@ def capture_timestamp_utc():
 
 
 def say(text, blocking=False):
-    # Check if mac, linux, or windows.
-    if platform.system() == "Darwin":
-        cmd = f'say "{text}"'
-        if not blocking:
-            cmd += " &"
-    elif platform.system() == "Linux":
-        cmd = f'spd-say "{text}"'
-        if blocking:
-            cmd += "  --wait"
-    elif platform.system() == "Windows":
-        # TODO(rcadene): Make blocking option work for Windows
-        cmd = (
-            'PowerShell -Command "Add-Type -AssemblyName System.Speech; '
-            f"(New-Object System.Speech.Synthesis.SpeechSynthesizer).Speak('{text}')\""
-        )
+    system = platform.system()
 
-    os.system(cmd)
+    if system == "Darwin":
+        cmd = ["say", text]
+
+    elif system == "Linux":
+        cmd = ["spd-say", text]
+        if blocking:
+            cmd.append("--wait")
+
+    elif system == "Windows":
+        cmd = [
+            "PowerShell",
+            "-Command",
+            "Add-Type -AssemblyName System.Speech; "
+            f"(New-Object System.Speech.Synthesis.SpeechSynthesizer).Speak('{text}')",
+        ]
+
+    else:
+        raise RuntimeError("Unsupported operating system for text-to-speech.")
+
+    if blocking:
+        subprocess.run(cmd, check=True)
+    else:
+        subprocess.Popen(cmd, creationflags=subprocess.CREATE_NO_WINDOW if system == "Windows" else 0)
 
 
 def log_say(text, play_sounds, blocking=False):

lerobot/scripts/eval.py

@@ -454,7 +454,7 @@ def _compile_episode_data(
 
 
 @parser.wrap()
-def eval(cfg: EvalPipelineConfig):
+def eval_main(cfg: EvalPipelineConfig):
     logging.info(pformat(asdict(cfg)))
 
     # Check device is available
@@ -499,4 +499,4 @@ def eval(cfg: EvalPipelineConfig):
 
 if __name__ == "__main__":
     init_logging()
-    eval()
+    eval_main()

lerobot/scripts/visualize_dataset_html.py

@@ -194,7 +194,7 @@ def run_server(
             ]
 
             response = requests.get(
-                f"https://huggingface.co/datasets/{repo_id}/resolve/main/meta/episodes.jsonl"
+                f"https://huggingface.co/datasets/{repo_id}/resolve/main/meta/episodes.jsonl", timeout=5
             )
             response.raise_for_status()
             # Split into lines and parse each line as JSON
@@ -318,7 +318,9 @@ def get_episode_language_instruction(dataset: LeRobotDataset, ep_index: int) ->
 
 
 def get_dataset_info(repo_id: str) -> IterableNamespace:
-    response = requests.get(f"https://huggingface.co/datasets/{repo_id}/resolve/main/meta/info.json")
+    response = requests.get(
+        f"https://huggingface.co/datasets/{repo_id}/resolve/main/meta/info.json", timeout=5
+    )
     response.raise_for_status()  # Raises an HTTPError for bad responses
     dataset_info = response.json()
     dataset_info["repo_id"] = repo_id

lerobot/templates/visualize_dataset_template.html

@@ -42,22 +42,22 @@
             <ul>
                 <template x-for="episode in paginatedEpisodes" :key="episode">
                     <li class="font-mono text-sm mt-0.5">
-                        <a :href="'episode_' + episode" 
+                        <a :href="'episode_' + episode"
                            :class="{'underline': true, 'font-bold -ml-1': episode == {{ episode_id }}}"
                            x-text="'Episode ' + episode"></a>
                     </li>
                 </template>
             </ul>
-        
+
             <div class="flex items-center mt-3 text-xs" x-show="totalPages > 1">
-                <button @click="prevPage()" 
+                <button @click="prevPage()"
                         class="px-2 py-1 bg-slate-800 rounded mr-2"
                         :class="{'opacity-50 cursor-not-allowed': page === 1}"
                         :disabled="page === 1">
                     &laquo; Prev
                 </button>
                 <span class="font-mono mr-2" x-text="` ${page} / ${totalPages}`"></span>
-                <button @click="nextPage()" 
+                <button @click="nextPage()"
                         class="px-2 py-1 bg-slate-800 rounded"
                         :class="{'opacity-50 cursor-not-allowed': page === totalPages}"
                         :disabled="page === totalPages">
@@ -65,10 +65,10 @@
                 </button>
             </div>
         </div>
-        
+
         <!-- episodes menu for small screens -->
         <div class="flex overflow-x-auto md:hidden" x-data="episodePagination">
-            <button @click="prevPage()" 
+            <button @click="prevPage()"
                     class="px-2 bg-slate-800 rounded mr-2"
                     :class="{'opacity-50 cursor-not-allowed': page === 1}"
                     :disabled="page === 1">&laquo;</button>
@@ -83,7 +83,7 @@
                     </p>
                 </template>
             </div>
-            <button @click="nextPage()" 
+            <button @click="nextPage()"
                     class="px-2 bg-slate-800 rounded ml-2"
                     :class="{'opacity-50 cursor-not-allowed': page === totalPages}"
                     :disabled="page === totalPages">&raquo; </button>
@@ -476,7 +476,7 @@
                 episodes: {{ episodes }},
                 pageSize: 100,
                 page: 1,
-                
+
                 init() {
                     // Find which page contains the current episode_id
                     const currentEpisodeId = {{ episode_id }};
@@ -485,23 +485,23 @@
                         this.page = Math.floor(episodeIndex / this.pageSize) + 1;
                     }
                 },
-                
+
                 get totalPages() {
                     return Math.ceil(this.episodes.length / this.pageSize);
                 },
-                
+
                 get paginatedEpisodes() {
                     const start = (this.page - 1) * this.pageSize;
                     const end = start + this.pageSize;
                     return this.episodes.slice(start, end);
                 },
-                
+
                 nextPage() {
                     if (this.page < this.totalPages) {
                         this.page++;
                     }
                 },
-                
+
                 prevPage() {
                     if (this.page > 1) {
                         this.page--;
@@ -515,7 +515,7 @@
         window.addEventListener('keydown', (e) => {
             // Use the space bar to play and pause, instead of default action (e.g. scrolling)
             const { keyCode, key } = e;
-            
+
             if (keyCode === 32 || key === ' ') {
                 e.preventDefault();
                 const btnPause = document.querySelector('[x-ref="btnPause"]');

pyproject.toml

@@ -48,7 +48,7 @@ dependencies = [
     "omegaconf>=2.3.0",
     "opencv-python>=4.9.0",
     "packaging>=24.2",
-    "pyav>=12.0.5",
+    "av>=12.0.5",
     "pymunk>=6.6.0",
     "pynput>=1.7.7",
     "pyzmq>=26.2.1",
@@ -111,10 +111,19 @@ exclude = [
     "venv",
 ]
 
-
 [tool.ruff.lint]
 select = ["E4", "E7", "E9", "F", "I", "N", "B", "C4", "SIM"]
 
+[tool.bandit]
+exclude_dirs = [
+    "tests",
+    "benchmarks",
+    "lerobot/common/datasets/push_dataset_to_hub",
+    "lerobot/common/datasets/v2/convert_dataset_v1_to_v2",
+    "lerobot/common/policies/pi0/conversion_scripts",
+    "lerobot/scripts/push_dataset_to_hub.py",
+]
+skips = ["B101", "B311", "B404", "B603"]
 
 [tool.typos]
 default.extend-ignore-re = [