admin/lerobot
1
0
Fork 0
mirror of https://github.com/huggingface/lerobot.git synced 2026-07-24 18:26:11 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(audio visualization): making audio visualization robust to multiple channel audio data

Browse Source
This commit is contained in:
CarolinePascal
2025-10-27 14:28:12 +01:00
parent 6bd40ca219
commit 1ee4d84f07
1 changed files with 19 additions and 16 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/lerobot/utils/visualization_utils.py
+6 -3
View File
@@ -152,8 +152,11 @@ def log_rerun_data(
for i, vi in enumerate(arr): for i, vi in enumerate(arr):
rr.log(f"{key}_{i}", rr.Scalars(float(vi))) rr.log(f"{key}_{i}", rr.Scalars(float(vi)))
elif arr.ndim == 2: elif arr.ndim == 2:
for i, channel_arr in enumerate(arr.T):
rr.send_columns( rr.send_columns(
"audio/" + key, "audio/"
+ key
+ f"_channel_{i}", # TODO(CarolinePascal): Get actual channel number/name
indexes=[ indexes=[
rr.TimeSecondsColumn( rr.TimeSecondsColumn(
"episode_time", "episode_time",
@@ -161,12 +164,12 @@ def log_rerun_data(
+ np.linspace( + np.linspace(
-DEFAULT_AUDIO_CHUNK_DURATION, -DEFAULT_AUDIO_CHUNK_DURATION,
0, 0,
len(observation[key]), len(channel_arr),
endpoint=False, endpoint=False,
), ),
) )
], ],
columns=rr.Scalar.columns(scalar=np.mean(observation[key], axis=1)), columns=rr.Scalar.columns(scalar=channel_arr),
) )
elif arr.ndim == 3: elif arr.ndim == 3:
rr.log(key, rr.Image(arr), static=True) rr.log(key, rr.Image(arr), static=True)