normalize
normalize: 70 implementations from 82 papers ran on one shared input (census 2026-09-22, battery b986f7e04d79); they produced 51 distinct outputs across 11 buckets, one shared input per bucket.
Identical values to six decimals (the recorded digest) on the shared input are agreement on those inputs, not equivalence. Implementations are compared only within one bucket, the positional (rank, kind, dtype) of each array argument; the argument name is not part of the key because the harness draws the shared array from (rank, kind) and casts it to the dtype, whatever the name; each bucket's shared input is fixed by that key, so members of one bucket saw bitwise-identical inputs under their own scalar arguments. A cluster is the set of members whose recorded output digest is identical. Nothing here says which computation a paper's method intended, and nothing reproduces a paper's results.
Not compared, and not in the tables or the counts above:
- 9 more implementations of this name were examined but did not run on the shared input (RuntimeError 5, ValueError 2, AttributeError 1, IndexError 1).
- 4 more implementations from 4 papers took no array argument and ran only on their own fixture arguments; no shared input existed for them, so they are not compared (on 3 the recorded shared-output digest is the own-fixture digest).
- Every compared output was digested.
- The harness also ran torch's own computation of this name under 15 named conventions; those rows are not implementations, are not counted, and only label a cluster whose digest they share.
Bucket 1 of 11: arg 1: rank 2, kind float, dtype float32
37 implementations from 42 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 23 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 9 implementations 12 papers 2803cfee2d9a recorded values differ by up to 2.98e-08 same digest as torch: l2_last_dim |
Transformer-ReID-Survey/UnTransReID_VI_ReID/clustercontrast/trainers.py e3b2a83f recorded x:2/float/float32; axis=-1
train_gta2cityscapes_multi.py ebaa2a39 recorded feature:2/float/float32 adapter/proceed.py 3a79b318 recorded W:2/float/float32; max_norm=1 bliss/models/models.py 5f809791 recorded arr:2/float/float32 tutorials/GAN_geometry_demo.py 754bece3 recorded vecs:2/float/float32; R=1 eval_knn.py d4ea389f recorded x:2/float/float32
method/network.py dcaf9c10 recorded angles:2/float/float32 otk/layers.py ea907204 recorded x:2/float/float32; p=2, dim=-1, inplace=True model.py f838567d recorded feat:2/float/float32 |
[0.60254, -0.512675, 0.159124, -0.230934, -0.301783, -0.344442, -0.221532, 0.191492, …] shape [4, 8] · float32 · Tensor |
| 3 implementations 3 papers 4171f3b5f627 recorded values differ by up to 2.98e-08 |
approximation_quality.py 0024f6e5 recorded x:2/float/float32 context_algs/ensemble_sampling.py 2148eda0 recorded matrix:2/float/float32 models/ts2vec/fsnet.py ab9d34a1 recorded W:2/float/float32 |
[0.316604, -0.269384, 0.0836115, -0.121344, -0.158572, -0.180987, -0.116404, 0.100619, …] shape [4, 8] · float32 · Tensor |
| 3 implementations 3 papers 771dbde58a2f recorded values identical same digest as torch: minmax_global |
preprocess/pancreas_preprocess.py 3398752e recorded data:2/float/float32 HyRA.py 6af53c0a recorded ten:2/float/float32
models.py e0d1f6fb recorded x:2/float/float32 |
[0.868319, 0.0225855, 0.53205, 0.236246, 0.182517, 0.150166, 0.243376, 0.556597, …] shape [4, 8] · float32 · Tensor |
| 2 implementations 3 papers 01d5f95c1450 recorded values identical |
RBFleX_NAS-Bench-201.py de4d734a recorded x:2/float/float32; axis=0 pcb-merging.py 3f0c220e recorded x:2/float/float32; dim=0 |
[1, 0.0345076, 0.650624, 0.33933, 0.2684, 0.0429106, 0.816733, 0.739644, …] shape [4, 8] · float32 · Tensor |
| 2 implementations 1 paper 52b2d2679944 recorded values identical |
fasth_wrapper.py 0e3494d3 recorded V:2/float/float32 fasthpp.py 5bd08a53 recorded V:2/float/float32 |
[2.99171, 3.1385, 2.37496, 2.82426] shape [4] · float32 · Tensor |
| 1 implementation 2 papers 3b0770b06397 |
reproduce_WinCLIP.py 0c3f9d9b recorded pred:2/float/float32; max_value=1.5, min_value=-1.0 |
[1.12105, -0.213511, 0.590421, 0.123645, 0.0388602, -0.0121885, 0.134896, 0.629156, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 2 papers ee20ec6baa9c |
preprocess_data.py 7baf753d recorded a:2/float/float32 |
[1, 0.0274511, 0.751886, 0.279741, 0.131726, 0.278117, 0.249376, 0.791199, …] shape [4, 8] · float32 · ndarray |
| 1 implementation 1 paper 0aca4709beba |
EvaluateClassicalModels/run_model_evaluation.py c848ffe7 recorded X:2/float/float32 |
[1.497, -0.948, 0.232, -0.3, -0.613, -0.918, 0.292, 0.331, …] shape [4, 8] · float32 · ndarray |
| 1 implementation 1 paper 0ef5f13e172b |
fskd.py e83e7d75 recorded v:2/float/float32 |
[0.289525, -0.246344, 0.0764601, -0.110965, -0.145009, -0.165507, -0.106448, 0.0920132, …] shape [4, 8] · float32 · ndarray |
| 1 implementation 1 paper 2e51512274a5 |
model_utilities.py ebcd07e9 recorded X:2/float/float32 |
[0.00706913, -0.00601481, 0.00186688, -0.00270936, -0.00354059, -0.00404106, -0.00259906, 0.00224662, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper 4004745eb8e5 same digest as torch: zscore_global |
ryRecog03.py 720014c5 recorded x:2/float/float32 |
[1.86047, -1.41588, 0.557774, -0.58816, -0.796307, -0.921632, -0.56054, 0.652867, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper 43129fb601cc |
scripts/eval_3dfront.py f89be01b recorded vertices:2/float/float32; scale=1 |
[1, -0.930985, 0.301248, -0.678992, 39.9593, -0.443765, 1.28745, 0.582397, …] shape [4, 8] · float32 · ndarray |
| 1 implementation 1 paper 48e680d0884b |
model/parametermodel.py ccd4ef08 recorded x:2/float/float32 |
[2.60526, -4.06755, -0.0478935, -2.38177, -2.8057, -3.06094, -2.32552, 0.145778, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper 57a439b58dcd |
utils.py 1b050e97 recorded t:2/float/float32 |
[-0.915424, 0.778893, -0.241753, 0.350851, 0.458492, 0.523301, 0.336568, -0.290929, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper 5cb0837f629c |
lab4d/render.py 99ee6315 recorded t:2/float/float32 |
[221.421, 5.75929, 135.673, 60.2428, 46.5418, 38.2924, 62.0609, 141.932, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper 84e1d0ae4d42 |
utils.py 748339aa recorded x:2/float/float32 |
[175, 90, 141, 112, 106, 103, 112, 144, …] shape [4, 8] · uint8 · ndarray |
| 1 implementation 1 paper 90af06b33d1c |
utils/diff_erank_single_sentence.py 34a7e871 recorded R:2/float/float32 |
[0.610934, -0.475295, 0.0487455, -0.121317, -0.127405, -0.598834, 0.0419297, 0.0846632, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper a74e972493b0 |
src/train_student.py aeb60ff8 recorded y:2/float/float32; norm_params={'max_val': 10.0, 'min_val': 2.0} |
[-0.0246714, -0.441722, -0.190493, -0.336361, -0.362856, -0.378809, -0.332845, -0.178389, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper bc59049858cd |
all_normalization_transformer/all_normalization_transformer.py 1d4fa3c9 recorded t:2/float/float32; eps=1e-08 |
[1.99163, -1.25172, 0.702057, -0.432336, -0.638387, -0.762449, -0.404994, 0.796192, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper d4b093c1926c |
methods.py 5eff8cd5 recorded x:2/float/float32 |
[0.602538, -0.512673, 0.159123, -0.230933, -0.301782, -0.344441, -0.221531, 0.191491, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper d7bd5ceec8a6 |
AdversarialTrainer.py 839a94c3 recorded tensor:2/float/float32; mean=[0.5, 0.3, 0.2, 0.1], std=[1.0, 1.5, 0.8, 1.2] |
[1.80263, -1.53378, 0.476053, -0.690887, -0.902849, -1.03047, -0.662761, 0.572889, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper d858bb1a4941 |
cnn/architect/architect_edarts.py 35516df4 recorded x:2/float/float32; dim=1, min_v=1e-05 |
[0.632142, 3.50678e-06, 0.166941, 3.50678e-06, 3.50678e-06, 3.50678e-06, 3.50678e-06, 0.200899, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper e063af5f422b |
RapidIn/calc_inner.py ec8f9379 recorded x:2/float/float32 |
[0.884225, -0.62132, 0.393433, -0.33966, -0.582611, -0.334824, -0.400334, 0.410058, …] shape [4, 8] · float32 · Tensor |
Bucket 2 of 11: arg 1: rank 4, kind float, dtype float32
9 implementations from 10 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 6 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 4 implementations 4 papers 6a027a3d1dea recorded values identical |
purification.py 3eec6934 recorded x:4/float/float32 Dirty_code_for_attack/Simsiam_backdoor_eval.py 5e6a53c9 recorded t:4/float/float32 ssp.py 80be5f7a recorded t:4/float/float32 Environment.py fb4cc72b recorded train_data:4/float/float32; config={'dataset': 'ImageNet', 'device': 'cpu'} |
[1.03976, 0.733971, 4.19688, 2.67921, 2.84845, -2.99095, -3.30344, 1.51519, …] shape [2, 3, 4, 4] · float32 · Tensor |
| 1 implementation 2 papers 1c3f932072ee |
models/Network.py 37f8b080 recorded images:4/float/float32 |
[0.44621, 0.306159, 1, 1, 1, -1, -1, 0.663958, …] shape [2, 3, 4, 4] · float32 · Tensor |
| 1 implementation 1 paper 2361e908de8c |
code/attacks.py dd42cbeb recorded X:4/float/float32 |
[0.899566, 0.63889, 3.59094, 2.29716, 2.44143, -2.53651, -2.8029, 1.30486, …] shape [2, 3, 4, 4] · float32 · Tensor |
| 1 implementation 1 paper 48bf8b9eb7d0 |
loc_loss.py 348efed2 recorded tensor:4/float/float32; eps=1e-08 |
[0.728997, 0.713934, 0.884524, 0.80976, 0.818097, 0.530436, 0.515043, 0.752418, …] shape [2, 3, 4, 4] · float32 · Tensor |
| 1 implementation 1 paper 50bb61297c67 |
robustness.py a8dfc8ec recorded t:4/float/float32 |
[1.14535, 0.799206, 4.71916, 3.00119, 3.19276, -3.41734, -3.77107, 1.68354, …] shape [2, 3, 4, 4] · float32 · Tensor |
| 1 implementation 1 paper 965b2781135f |
src/tinyedm/networks.py 2add1f3e recorded x:4/float/float32; eps=0.0001 |
[0.758631, 0.685165, 1.51713, 1.15251, 1.19317, -0.20975, -0.284825, 0.872854, …] shape [2, 3, 4, 4] · float32 · Tensor |
Bucket 3 of 11: arg 1: rank 1, kind float, dtype float64
6 implementations from 8 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 5 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 2 implementations 4 papers 7d4be19aa334 recorded values identical |
utils/mpi_utils.py 1d6f8390 recorded x:1/float/float64 one code sha held from 2 papers' repositories models/sem_mapping.py ec5d439a recorded v:1/float/float64 |
[-0.40645, 0.070657, 0.403627, -0.342735, -0.0792565, 0.0238467, -0.732257, -0.079831] shape [8] · float64 · ndarray |
| 1 implementation 1 paper 65c5345153d2 |
inference/candidate_selection.py f9308298 recorded weights:1/float/float64 |
[0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125, 0.125] shape [8] · float64 · ndarray |
| 1 implementation 1 paper 7912df950dee |
maxoutCNN.py 5d33bc8a recorded data:1/float/float64 |
[-0.34103, 0.276105, 0.7068, -0.258615, 0.082193, 0.215556, -0.76246, 0.0814498] shape [8] · float64 · ndarray |
| 1 implementation 1 paper 88b5149f24c2 |
evaluate/sim.py 53e96986 recorded v:1/float/float64; n=[2.5, 0.5] |
[-5, -4.68574, -3.20478, -5, -5, -4.89394, -5, -5] shape [8] · float64 · ndarray |
| 1 implementation 1 paper ef212641461a |
generation_scripts/grammars/cogs-preprocess.py 45fed9ff recorded probs:1/float/float64 |
[-0.461322, 0.599699, 1.34018, -0.319628, 0.266312, 0.495599, -1.18587, 0.265034] shape [8] · float64 · ndarray |
Bucket 4 of 11: arg 1: rank 2, kind float, dtype float64
5 implementations from 5 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 5 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper 2803cfee2d9a |
retrieval/deep_query.py 3c9e01b7 recorded x:2/float/float64; copy=False |
[0.60254, -0.512675, 0.159124, -0.230934, -0.301783, -0.344442, -0.221532, 0.191492, …] shape [4, 8] · float64 · ndarray |
| 1 implementation 1 paper 7c252ed0653b |
NUTMEG/nutmeg.py 44ff7c0f recorded x:2/float/float64; smoothing=0.1 |
[-1.62733, 1.22632, -0.492701, 0.505389, 0.686681, 0.795837, 0.481332, -0.575525, …] shape [4, 8] · float64 · ndarray |
| 1 implementation 1 paper ab42e90e344c |
src/lshot_update.py 5875172e recorded Y_in:2/float/float64 |
[0.530059, 0.0188513, 0.14067, 0.043793, 0.0354283, 0.0311835, 0.0450422, 0.154973, …] shape [4, 8] · float64 · ndarray |
| 1 implementation 1 paper bc31536e7790 |
pre-processing/homography_align_SIFT.py 34900df2 recorded x:2/float/float64; lower=0, upper=255 |
[221.421, 5.75929, 135.673, 60.2428, 46.5418, 38.2924, 62.0609, 141.932, …] shape [4, 8] · float64 · ndarray |
| 1 implementation 1 paper fcca2a21c96c |
utils/data_preprocess.py c5353d6f recorded arr:2/float/float64 |
[1.89024, -1.43853, 0.566699, -0.597571, -0.809049, -0.936379, -0.56951, 0.663313, …] shape [4, 8] · float64 · ndarray |
Bucket 5 of 11: arg 1: rank 3, kind float, dtype float32
4 implementations from 6 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 4 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 2 papers 074d4e3a6ff9 |
one code sha held from 2 papers' repositories iCT/cm/unet.py b0e4dda4 recorded x:3/float/float32; dim=None, eps=0.0001 |
[-0.906764, 1.70855, -1.27802, -0.262148, 0.153301, -1.45667, -0.345339, -0.386945, …] shape [2, 4, 8] · float32 · Tensor |
| 1 implementation 2 papers 7f7b8d46b734 |
train/train_internvl.py 52e1d206 recorded logit:3/float/float32 |
[-0.564221, 2.07076, -0.938274, 0.0852428, 0.503817, -1.11826, 0.00142618, -0.0404925, …] shape [2, 4, 8] · float32 · Tensor |
| 1 implementation 1 paper 2795134bc599 |
code/gen_eval.py 3cabbd74 recorded img:3/float/float32 |
[-2.78625, 2.36571, -3.51761, -1.51641, -0.698008, -3.86952, -1.68029, -1.76225, …] shape [2, 4, 8] · float32 · Tensor |
| 1 implementation 1 paper 49738fec43f3 |
tiny-yolo2-reduced.py d989af7b recorded image:3/float/float32 |
[-0.00350246, 0.00659942, -0.00493649, -0.00101257, 0.000592141, -0.00562651, -0.0013339, -0.00149461, …] shape [2, 4, 8] · float32 · Tensor |
Bucket 6 of 11: arg 1: rank 1, kind float, dtype float32
2 implementations from 2 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 2 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper 7d4be19aa334 same digest as torch: l2_last_dim |
easyeditor/models/wise/WISE.py a6e2f5aa recorded v:1/float/float32; eps=1e-06 |
[-0.40645, 0.070657, 0.403627, -0.342735, -0.0792565, 0.0238467, -0.732257, -0.079831] shape [8] · float32 · ndarray |
| 1 implementation 1 paper ebfdf9421e6b same digest as torch: minmax_global, minmax_last_dim |
self_driving.py be165158 recorded X:1/float/float32 |
[0.286831, 0.706863, 1, 0.342924, 0.574883, 0.665652, 0, 0.574377] shape [8] · float32 · Tensor |
Bucket 7 of 11: arg 1: rank 2, kind float, dtype float32 · arg 2: rank 1, kind float, dtype float32 · arg 3: rank 1, kind float, dtype float32
2 implementations from 3 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 2 distinct outputs, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 2 papers f8598acfe2e1 |
cadm/dynamics/mlp_cadm_ensemble_cem_dynamics.py a7c346d6 recorded data_array:2/float/float32, mean:1/float/float32, std:1/float/float32 |
[-2.9943, -10.7611, -0.469645, -0.0935567, 4.12239, -20.4312, -0.593009, -4.22694, …] shape [4, 8] · float32 · Tensor |
| 1 implementation 1 paper ce8c2075b4f1 |
DiGN.py 12e0fe0c recorded tensor:2/float/float32, mean:1/float/float32, std:1/float/float32 |
[-2.9943, 0.696864, -1.52667, -0.235651, -0.00115004, 0.140042, -0.266767, -1.6338, …] shape [4, 8] · float32 · Tensor |
Bucket 8 of 11: arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32 · arg 3: rank 2, kind float, dtype float32
2 implementations from 3 papers share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 2 implementations 3 papers 97656392209b recorded values identical |
utils/loss_utils.py e3442ad3 recorded input:2/float/float32, mean:2/float/float32, std:2/float/float32 bayes_race/raceline/generate_raceline_ethz.py 348378ff recorded y_eval:2/float/float32, mean_y:2/float/float32, std_y:2/float/float32 |
[0, -0, 0, -0, -0, -0, -0, 0, …] shape [4, 8] · float32 · Tensor |
Bucket 9 of 11: arg 1: rank 2, kind float, dtype float32 · arg 2: rank 2, kind float, dtype float32
1 implementation from 1 paper share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper 2f11afb1a7e1 |
models/losses.py b0bf73a5 recorded loss:2/float/float32, weight:2/float/float32 |
[2.61015, 6.09388, 1.0056, 4.13737, 2.40145, 6.83858, 2.74075, 1.29008] shape [8] · float32 · Tensor |
Bucket 10 of 11: arg 1: rank 2, kind int, dtype uint8
1 implementation from 1 paper share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper 4bb18a7273bd |
test_new.py b58ae1ca recorded input_data:2/int/uint8 |
[-0.984314, -0.976471, -0.984314, -0.992157, -0.992157, -0.976471, -1, -0.992157, …] shape [4, 8] · float32 · ndarray |
Bucket 11 of 11: arg 1: rank 3, kind int, dtype uint8
1 implementation from 1 paper share this bucket (rank, kind, dtype of each array argument, positional; each member's recorded signature, argument name included, is shown under it); 1 distinct output, largest cluster first. Values are the first 8 of the recorded output, flattened.
| Cluster (same digest to six decimals) | Members | Shared output on this bucket's input |
|---|---|---|
| 1 implementation 1 paper ea3862f595b8 |
testing/test_pose.py 9d8d1b76 recorded origin_img:3/int/uint8 |
[-0.492188, -0.5, -0.496094, -0.488281, -0.492188, -0.496094, -0.496094, -0.492188, …] shape [2, 4, 8] · float32 · ndarray |
Identical values to six decimals (the recorded digest) on the shared input are agreement on those inputs, not equivalence; where a cluster's members carry recorded values, the largest difference among them is shown under the cluster. Paper titles are the archive's archive 2025-07-28 where the paper is in the archive and the graph's where it was added by Syntology; papers with no page here are shown by their recorded paper id only. A paper count above the implementation count means one implementation (one code sha) is held from several papers' repositories and counts once. Per-sample status, licence and fingerprint records for each paper are on its paper page. JSON twin: /census/normalize.json.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections