admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-07-07 18:11:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(remove ffv1): removing ffv1 as it does not support MP4

Browse Source
This commit is contained in:
CarolinePascal
2026-06-24 22:02:51 +02:00
parent 5a7968b835
commit 96c1416bc9
7 changed files with 11 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/source/using_dataset_tools.mdx
+2 -2
View File
@@ -181,8 +181,8 @@ lerobot-edit-dataset \
lerobot-edit-dataset \
--repo_id lerobot/pusht_depth \
--operation.type reencode_videos \
--operation.rgb_encoder.vcodec libx264 \
--operation.depth_encoder.vcodec ffv1
--operation.rgb_encoder.vcodec h264 \
--operation.depth_encoder.crf 50
```
**Parameters:**
docs/source/video_encoding_parameters.mdx
+1 -1
View File
@@ -102,7 +102,7 @@ lerobot-record \
| Parameter | Type | Default | Description |
| ----------- | ------- | ------------ | --------------------------------------------------------------------------------------------------------------------------------- |
| `vcodec` | `str` | `"hevc"` | Defaults to HEVC Main 12 (a 12-bit-capable codec). `ffv1` is a lossless alternative. |
| `vcodec` | `str` | `"hevc"` | Defaults to HEVC Main 12 (a 12-bit-capable codec). For a lossless depth stream, keep HEVC and pass `extra_options={"x265-params": "lossless=1"}` (stays MP4-compatible). |
| `pix_fmt` | `str` | `"gray12le"` | Single-channel 12-bit pixel format used to carry the quantized codes. |
| `depth_min` | `float` | `0.01` | Depth in metres mapped to quantum `0`. Values below are clipped on decode. |
| `depth_max` | `float` | `10.0` | Depth in metres mapped to quantum `4095`. Values above are clipped on decode. |
src/lerobot/configs/video.py
+1 -5
View File
@@ -37,7 +37,7 @@ HW_VIDEO_CODECS = [
"h264_qsv", # Intel Quick Sync
]
VALID_VIDEO_CODECS: frozenset[str] = frozenset(
{"h264", "hevc", "libsvtav1", "ffv1", "auto", *HW_VIDEO_CODECS}
{"h264", "hevc", "libsvtav1", "auto", *HW_VIDEO_CODECS}
)
# Aliases for legacy video codec names.
VIDEO_CODECS_ALIASES: dict[str, str] = {"av1": "libsvtav1"}
@@ -234,10 +234,6 @@ class VideoEncoderConfig:
elif self.vcodec == "h264_qsv":
set_if("global_quality", self.crf)
set_if("preset", self.preset)
elif self.vcodec == "ffv1":
# Lossless intra-frame codec. ``crf``/``preset``/``fast_decode``
# are not meaningful.
set_if("threads", encoder_threads)
else:
set_if("crf", self.crf)
set_if("preset", self.preset)
src/lerobot/datasets/depth_utils.py
+1 -1
View File
@@ -79,7 +79,7 @@ def quantize_depth(
Depth maps are packed into 12-bit integer frames so they fit in standard
high-bit-depth pixel formats (e.g. ``yuv420p12le`` / ``gray12le``)
and can be encoded by widely supported video codecs (HEVC Main 12, ffv1).
and can be encoded by widely supported video codecs (e.g. HEVC Main 12).
Logarithmic quantization is the default because it allocates more quanta
to near-range depth, which matches the (1/depth) error profile of typical
depth sensors. Math is ported from BEHAVIOR-1K's ``obs_utils.py``.
src/lerobot/scripts/lerobot_edit_dataset.py
+2 -2
View File
@@ -224,8 +224,8 @@ Re-encode both RGB and depth videos in a dataset (depth quantization params are
lerobot-edit-dataset \
--repo_id lerobot/pusht_depth \
--operation.type reencode_videos \
--operation.rgb_encoder.vcodec libx264 \
--operation.depth_encoder.vcodec ffv1
--operation.rgb_encoder.vcodec h264 \
--operation.depth_encoder.extra_options '{"x265-params": "lossless=1"}'
Using JSON config file:
lerobot-edit-dataset \
tests/datasets/test_datasets.py
+1 -2
View File
@@ -1531,7 +1531,6 @@ def test_valid_video_codecs_constant():
assert "h264" in VALID_VIDEO_CODECS
assert "hevc" in VALID_VIDEO_CODECS
assert "libsvtav1" in VALID_VIDEO_CODECS
assert "ffv1" in VALID_VIDEO_CODECS
assert "auto" in VALID_VIDEO_CODECS
assert "h264_videotoolbox" in VALID_VIDEO_CODECS
assert "h264_nvenc" in VALID_VIDEO_CODECS
@@ -1539,7 +1538,7 @@ def test_valid_video_codecs_constant():
assert "h264_qsv" in VALID_VIDEO_CODECS
assert "hevc_videotoolbox" in VALID_VIDEO_CODECS
assert "hevc_nvenc" in VALID_VIDEO_CODECS
assert len(VALID_VIDEO_CODECS) == 11
assert len(VALID_VIDEO_CODECS) == 10
def test_delta_timestamps_with_episodes_filter(tmp_path, empty_lerobot_dataset_factory):
tests/scripts/test_edit_dataset_parsing.py
+3 -3
View File
@@ -123,15 +123,15 @@ class TestDepthEncoderParsing:
"test/repo",
"--operation.type",
"reencode_videos",
"--operation.depth_encoder.vcodec",
"ffv1",
"--operation.depth_encoder.extra_options",
'{"x265-params": "lossless=1"}',
"--operation.depth_encoder.depth_max",
"12.0",
"--operation.depth_encoder.use_log",
"false",
]
)
assert cfg.operation.depth_encoder.vcodec == "ffv1"
assert cfg.operation.depth_encoder.extra_options == {"x265-params": "lossless=1"}
assert cfg.operation.depth_encoder.depth_max == 12.0
assert cfg.operation.depth_encoder.use_log is False