Download weights/instances/i36df1f8f/task.json from Boom5426/WeightCode: direct link, hf CLI and curl.
- Browser
- Download file 4.45 kB
-
https://huggingface.co/datasets/Boom5426/WeightCode/resolve/main/weights/instances/i36df1f8f/task.json
- Command line
-
hf download hf://datasets/Boom5426/WeightCode/weights/instances/i36df1f8f/task.json
-
curl -L -o task.json https://huggingface.co/datasets/Boom5426/WeightCode/resolve/main/weights/instances/i36df1f8f/task.json
4.45 kB
| { | |
| "classes": [ | |
| "airplane", | |
| "automobile", | |
| "bird", | |
| "cat", | |
| "deer", | |
| "dog", | |
| "frog", | |
| "horse", | |
| "ship", | |
| "truck" | |
| ], | |
| "evidence": { | |
| "fail": { | |
| "count": 16, | |
| "description": "examples from the deployment data that the classifier gets wrong, drawn at random from all of its errors there", | |
| "keys": [ | |
| "x_fail", | |
| "y_fail", | |
| "pred_fail" | |
| ] | |
| }, | |
| "ids": { | |
| "description": "one opaque id per example, fail examples first", | |
| "keys": [ | |
| "split_ids" | |
| ] | |
| }, | |
| "random": { | |
| "count": 48, | |
| "description": "random examples from the same deployment data, whether the classifier gets them right or not", | |
| "keys": [ | |
| "x_rand", | |
| "y_rand", | |
| "pred_rand" | |
| ] | |
| } | |
| }, | |
| "limit_definitions": { | |
| "max_changed_frac": "changed entries over all parameter entries; an entry is changed if its float32 value differs from the parent's; null: this limit is not applied", | |
| "max_edit_frac": "numbers the patch specifies, counted per op (add_dense: tensor size; index-form set/add: number of values; slice-form set/add: 1 with one value, the block size with a values file; scale_rows/cols: positions; project_out: direction length; low_rank_add: r*(rows+cols); tie: group-map length), over all parameter entries", | |
| "max_file_bytes": "total bytes of the distinct .npy files one patch references", | |
| "max_ops": "operations per patch", | |
| "max_rank": "rank r of u and v in low_rank_add", | |
| "max_rel_frob": "per tensor of rank >= 2, ||child - parent||_F / ||parent||_F; 1-D tensors are exempt", | |
| "rel_frob_min_ndim": "smallest tensor rank that max_rel_frob applies to; tensors of lower rank are exempt" | |
| }, | |
| "limits": { | |
| "max_changed_frac": null, | |
| "max_edit_frac": 0.05, | |
| "max_file_bytes": 67108864, | |
| "max_ops": 64, | |
| "max_rank": 8, | |
| "max_rel_frob": 0.5, | |
| "rel_frob_min_ndim": 2 | |
| }, | |
| "patch_check": { | |
| "budget_queries": 8, | |
| "final_selection": "kept at the end: among the checked patches and the unedited classifier (S = 0), the one with the highest S on the check split; ties go to fewer changed parameters, then to the earlier query; a patch with S <= 0 and an unchecked patch are never kept", | |
| "invalid_patches": "a patch that fails validation consumes no query", | |
| "readout": "each query applies one patch and returns aggregate numbers on a held-out check split of the deployment data, each against the unedited classifier: accuracy, fixed (images the unedited classifier got wrong that the patched one gets right), broken (images the unedited classifier got right that the patched one gets wrong), the changed-parameter count and S = fixed - broken; no per-image information" | |
| }, | |
| "patch_format": { | |
| "allowed_ops": [ | |
| "scale_rows", | |
| "scale_cols", | |
| "set", | |
| "add", | |
| "project_out", | |
| "low_rank_add", | |
| "add_dense", | |
| "tie" | |
| ], | |
| "schema": "wrsi.patch.v0" | |
| }, | |
| "schema": "wrsi.task.v1", | |
| "statement": "Improve a trained image classifier on the data it is deployed on. The evidence file holds examples from the deployment data that the classifier gets wrong, plus random examples from the same data, with their labels and the classifier's predictions. Change the classifier's weights directly, without training it, so that it classifies more new deployment images correctly, while the images it already classifies correctly stay correct.", | |
| "training_free_rule": { | |
| "allowed": [ | |
| "forward passes, activations and weight inspection", | |
| "label-conditioned descriptive statistics (class means and their differences, SVD/PCA, correlations)", | |
| "single-element ablation sweeps for attribution", | |
| "comparing up to 8 explicit candidate patches on evidence" | |
| ], | |
| "audit": "every executed cell is logged and audited against these rules", | |
| "forbidden": [ | |
| "gradients of any loss with respect to parameters (autograd, hand-written backpropagation, finite-difference gradients)", | |
| "solvers whose targets derive from evidence labels (lstsq, solve, pinv, inv, closed-form regression or a linear probe)", | |
| "any .fit( call", | |
| "iterative parameter updates driven by a label-dependent objective (gradient descent, coordinate descent, evolutionary or random search that keeps the best)", | |
| "scoring more than 8 distinct full candidate patches against evidence labels in one session" | |
| ] | |
| } | |
| } | |