knot
Datasets
All datasets matching “knot”forceflow-bspline-knot-analysis
ForceFlow Adaptive B-spline Knot Analysis
Training-free adaptive cubic B-spline fitting results on episode 0 of
JokerESC/ForceFlow.
No policy, SmolVLA model, or GPU was used.
Question
When action changes little but force/torque changes substantially, does fitting
one vector-valued B-spline to [action, force, torque] allocate knots near the
force event that are absent from an action-only fit?
Both fits use one common knot vector per multichannel trajectory. For the… See the full description on the dataset page: https://huggingface.co/datasets/sungkyunner/forceflow-bspline-knot-analysis.foldingdiff_knots
Datasets for training Knot vs Unknot classifier
The dataset contains several folders with PDB files for training the Knot vs Unknot classifier. Each folder contains a mixture of knotted and unknotted protein structures,
knotted structures start with 'k', unknotted with 'u'.
All structures are max. 128 AA long.
AFdataset - structures predicted from AlphaFold, based on Knot or Not paper
Artifical_knots_dataset - structures designed with RFdiffusion and Evodiff, based on "Unveiling… See the full description on the dataset page: https://huggingface.co/datasets/EvaKlimentova/foldingdiff_knots.univtac-tactile-knot-exchange-20episodes
UniVTAC: five episodes per task — action/tactile knot exchange
20 episodes (IDs 0–4 for four tasks), 2 budget regimes, 4 allocations each: 160 fitted action splines, no policy training. The question is whether tactile-selected timing can be incorporated at fixed action representation capacity with reproducible reconstruction benefit, not whether physical steering succeeds.
Bottom line
The current tactile-residual knot-pinning heuristic is NOT consistently better… See the full description on the dataset page: https://huggingface.co/datasets/sungkyunner/univtac-tactile-knot-exchange-20episodes.forceflow-dual-tolerance-knot-analysis
ForceFlow Action–Torque Dual-Tolerance B-spline Analysis
Training-free, CPU-only adaptive cubic B-spline results on
JokerESC/ForceFlow,
episode 0, event 1388 (steps 1348–1428). No policy or neural-network model was
constructed or executed.
Exact experiment rule
Action and torque reconstruction errors are kept separate. For each tolerance
value in 0.25, 0.50, 0.75, 1.00, 1.25:
The action global RMSE tolerance and torque global RMSE tolerance are set to
the same… See the full description on the dataset page: https://huggingface.co/datasets/sungkyunner/forceflow-dual-tolerance-knot-analysis.univtac-bspline-knot-analysis
UniVTAC × B-spline Policy fitting pilot
This repository fits cubic B-splines to an augmented target containing UniVTAC's next 8-D joint action and next 4,800-D GelSight Mini marker displacement. It deliberately performs fitting only, not policy training.
The completed pilot outputs are in results/. Each task/setting has:
a PNG of the complete episode with natural-language phase labels and action/tactile knot colors;
an NPZ containing the complete knot vector and every 4,808-D… See the full description on the dataset page: https://huggingface.co/datasets/sungkyunner/univtac-bspline-knot-analysis.knot-crossings
🪢 Knot Crossing Dataset
This dataset contains synthetic and real images of knots labeled by their number of crossings.It is used to train and evaluate models that classify a knot diagram into its crossing number.
📂 Dataset Structure
The dataset is organized into two splits:
train/
0/
3/
4/
...
test/
0/
3/
4/
...
Each folder corresponds to a crossing number class.
For example, 3/ contains diagrams of prime knots with 3 crossings.
👩🏻🏫 Classes
Each… See the full description on the dataset page: https://huggingface.co/datasets/tr33hugg3r/knot-crossings.