Instructions to use HannesVonEssen/microduck-backflip with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Microduck
How to use HannesVonEssen/microduck-backflip with Microduck:
# Replace SLOT with the slot specified in the model card (walk, stand, sitstand, ground_pick, kick_left, kick_right, roulade). sudo robotctl policy load SLOT HannesVonEssen/microduck-backflip
- Notebooks
- Google Colab
- Kaggle
Use task-specific hardware status and backflip warning
Browse files- README.md +1 -2
- SHA256SUMS +3 -3
- TRAINING.md +1 -1
- manifest.json +1 -1
README.md
CHANGED
|
@@ -9,8 +9,7 @@ thumbnail: https://huggingface.co/HannesVonEssen/microduck-backflip/resolve/main
|
|
| 9 |
|
| 10 |
<video src="https://huggingface.co/HannesVonEssen/microduck-backflip/resolve/main/media/preview.mp4" controls muted playsinline width="720"></video>
|
| 11 |
|
| 12 |
-
**Simulation only.
|
| 13 |
-
simulated recovery does not establish a safe physical landing.**
|
| 14 |
|
| 15 |
A backflip from a raised platform onto a soft-contact simulated mat. This is
|
| 16 |
not a standing-height backflip and has not been validated on hardware.
|
|
|
|
| 9 |
|
| 10 |
<video src="https://huggingface.co/HannesVonEssen/microduck-backflip/resolve/main/media/preview.mp4" controls muted playsinline width="720"></video>
|
| 11 |
|
| 12 |
+
**⚠️ Landings may break the robot.** Simulation only; not yet validated on hardware. Successful simulated recovery does not establish a safe physical landing.
|
|
|
|
| 13 |
|
| 14 |
A backflip from a raised platform onto a soft-contact simulated mat. This is
|
| 15 |
not a standing-height backflip and has not been validated on hardware.
|
SHA256SUMS
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
2d3fd9ffffa70a4307498ec630b14a6b4731e010f8ddda92686eaf65733c2dfe LICENSE
|
| 2 |
-
|
| 3 |
-
|
| 4 |
f7e523c19de5cd5e2ced430f3d9bb46bef1ed69ad0003cb45955d56169929cdd checkpoint.pt
|
| 5 |
b4ef411ee0c1357763390596beb8cd8c4cd64d0cbe98c11b9d7daf8e0371727b config.json
|
| 6 |
e5b5bfb4e5fe8320584a20d12d729081a8c29980919be9540e8256d0a3d0fee5 eval/backflip.json
|
| 7 |
d2478a39f6d03018d830be78c9d72a7322e755932f21e0c307305220ab890396 eval/onnx-parity.json
|
| 8 |
8e3b3fc4530883b403c5f6bec48371e04985b6a3c3957a02bcedd99896c38ba3 eval/render-20260929.json
|
| 9 |
a0c92f27287602cd7983d16957890584ce8b9d323a10ffc1aad1b1a60283a52a eval/summary.json
|
| 10 |
-
|
| 11 |
bfe4a0ff24290751bafe65ecb63d75a3acd02c7d4c05f09925c0f8c15e89a186 media/preview.mp4
|
| 12 |
841e4eebfbe9946145f88cf01ce3f8f8a501b6b9312b0f037f2b4e0a5f91a48c media/social-preview.jpg
|
| 13 |
b4d780ee2a2bf7aab5b33358124fd079b17a329412d496574296919dbc5b6ee0 media/social-preview.png
|
|
|
|
| 1 |
2d3fd9ffffa70a4307498ec630b14a6b4731e010f8ddda92686eaf65733c2dfe LICENSE
|
| 2 |
+
7b28df0ee3dca8f0fe35d575a2e68b11a091fdb6bd46c03a4782bc842770e996 README.md
|
| 3 |
+
19a2ee5d1143ef7a68d1022ec1004110fc307605105d50ebbb913ca8ba599633 TRAINING.md
|
| 4 |
f7e523c19de5cd5e2ced430f3d9bb46bef1ed69ad0003cb45955d56169929cdd checkpoint.pt
|
| 5 |
b4ef411ee0c1357763390596beb8cd8c4cd64d0cbe98c11b9d7daf8e0371727b config.json
|
| 6 |
e5b5bfb4e5fe8320584a20d12d729081a8c29980919be9540e8256d0a3d0fee5 eval/backflip.json
|
| 7 |
d2478a39f6d03018d830be78c9d72a7322e755932f21e0c307305220ab890396 eval/onnx-parity.json
|
| 8 |
8e3b3fc4530883b403c5f6bec48371e04985b6a3c3957a02bcedd99896c38ba3 eval/render-20260929.json
|
| 9 |
a0c92f27287602cd7983d16957890584ce8b9d323a10ffc1aad1b1a60283a52a eval/summary.json
|
| 10 |
+
c0d0d5b5c18a2a3689deca7cd8b46473a14f5456cebd3ecf7d46e36171cefef0 manifest.json
|
| 11 |
bfe4a0ff24290751bafe65ecb63d75a3acd02c7d4c05f09925c0f8c15e89a186 media/preview.mp4
|
| 12 |
841e4eebfbe9946145f88cf01ce3f8f8a501b6b9312b0f037f2b4e0a5f91a48c media/social-preview.jpg
|
| 13 |
b4d780ee2a2bf7aab5b33358124fd079b17a329412d496574296919dbc5b6ee0 media/social-preview.png
|
TRAINING.md
CHANGED
|
@@ -71,7 +71,7 @@ run YAML and hashes accompany the applicable HF training packages. Changes to
|
|
| 71 |
curricula/rewards beyond these recipes are new experiments and should be logged.
|
| 72 |
The training source's contact budgets include publication fixes.
|
| 73 |
|
| 74 |
-
**Simulation only.
|
| 75 |
continuation, evaluate complete physical maneuvers and resets with the correct
|
| 76 |
command slots and action bounds, not reward totals alone. The original montage's
|
| 77 |
exact checkpoint/seed mapping remains unverified; training documentation does
|
|
|
|
| 71 |
curricula/rewards beyond these recipes are new experiments and should be logged.
|
| 72 |
The training source's contact budgets include publication fixes.
|
| 73 |
|
| 74 |
+
**Simulation only; not yet validated on hardware.** For the backflip: **⚠️ Landings may break the robot.** After
|
| 75 |
continuation, evaluate complete physical maneuvers and resets with the correct
|
| 76 |
command slots and action bounds, not reward totals alone. The original montage's
|
| 77 |
exact checkpoint/seed mapping remains unverified; training documentation does
|
manifest.json
CHANGED
|
@@ -54,7 +54,7 @@
|
|
| 54 |
"onnx_export_path": "scripts/export.py",
|
| 55 |
"simulation_contact_capacity": 200,
|
| 56 |
"source_note": "Public two-policy split; original task physics and per-policy continuation settings unchanged from validated preparation. Pillar release excluded.",
|
| 57 |
-
"warning": "
|
| 58 |
"training_continuation": {
|
| 59 |
"repository": "Vottivott/microduck-playground",
|
| 60 |
"branch": "main",
|
|
|
|
| 54 |
"onnx_export_path": "scripts/export.py",
|
| 55 |
"simulation_contact_capacity": 200,
|
| 56 |
"source_note": "Public two-policy split; original task physics and per-policy continuation settings unchanged from validated preparation. Pillar release excluded.",
|
| 57 |
+
"warning": "⚠️ Landings may break the robot. Simulation only; not yet validated on hardware.",
|
| 58 |
"training_continuation": {
|
| 59 |
"repository": "Vottivott/microduck-playground",
|
| 60 |
"branch": "main",
|