{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/census/l2-norm","entry":"l2_norm","source":"Syntology differential census (groundwork/55, run v2_2026-09-22), per sample; not an archive number","census_date":"2026-09-22","battery_sha256":["b986f7e04d794a0d88ad4c5f32cf63ec3590b5deff0192150737bc5f1c0b4677"],"runner_sha256":["5a452d0e7c0da5b80771d1be2afe3572e253568cf5c5d59a0d08ebd663d00808"],"bucket_key":"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","bucket_fields":["rank","kind","dtype"],"claim":"Implementations sharing this entry name were each run on one shared input fixed by the positional (rank, kind, dtype) of their array arguments (the bucket). A cluster is the set whose recorded output digest (sha256 of the output rounded to 6 decimals) is identical. Identical values to six decimals on the shared input are agreement on those inputs, not a statement about the whole domain and not a substitution claim.","n_implementations_compared":17,"n_papers":25,"n_buckets":3,"n_distinct_outputs":8,"n_class_bearing":0,"not_compared":{"not_run_on_shared_input":{"n":0,"by_error":{}},"no_array_argument_ran_on_own_fixture_arguments_only":{"n":0,"n_papers":0,"recorded_shared_digest_equals_own_fixture_digest":0},"output_not_digested_non_numeric":{"n":0,"by_type":{}}},"withdrawn_excluded":0,"code_page":"/code/l2-norm","buckets":[{"bucket":[[2,"float","float32"]],"n_implementations_compared":11,"n_papers":17,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"2803cfee2d9ac85e","size":8,"n_papers":12,"shape":[4,8],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":8,"torch_reference_conventions_with_this_digest":["last_dim"],"values":[0.6025403738021851,-0.5126747488975525,0.15912389755249023,-0.23093350231647491,-0.30178332328796387,-0.3444417119026184,-0.2215321660041809,0.191491961479187,0.16532593965530396,0.30560609698295593,-0.12655700743198395,-0.49971216917037964,-0.09193398058414459,0.7398778200149536,-0.16402213275432587,0.175411194562912,0.027956431731581688,-0.6833271980285645,0.18194462358951569,-0.17102842032909393,-0.009513513185083866,-0.49717968702316284,-0.22605349123477936,0.41418537497520447,-0.28170618414878845,0.153411403298378,0.3345924913883209,0.3602772355079651,-0.43401941657066345,0.451168030500412,-0.46816498041152954,-0.2103588581085205],"values_recorded":32,"members":[{"code_sha256_prefix":"c54fea429589425d","path":"models/arcface_model.py","papers":["2505.19328","2509.20177","1905.11946","1912.13457","2112.05142"],"paper_pages":[{"arxiv_id":"2505.19328","page":"/paper/bah-dataset-for-ambivalence-hesitancy"},{"arxiv_id":"2509.20177","page":"/paper/arxiv-2509-20177"},{"arxiv_id":"1905.11946","page":"/paper/efficientnet-rethinking-model-scaling-for"},{"arxiv_id":"1912.13457","page":"/paper/faceshifter-towards-high-fidelity-and"},{"arxiv_id":"2112.05142","page":"/paper/hairclip-design-your-hair-by-text-and"}],"arg_sig_recorded":[["input",2,"float","float32"]],"scalar_args":{"axis":"1"},"class_bearing":false},{"code_sha256_prefix":"45837478fdb29f78","path":"losses.py","papers":["2307.10943","2408.13492"],"paper_pages":[{"arxiv_id":"2307.10943","page":"/paper/proxy-anchor-based-unsupervised-learning-for"},{"arxiv_id":"2408.13492","page":"/paper/online-continuous-generalized-category"}],"arg_sig_recorded":[["input",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"265995c2541317fa","path":"recognition/torchkit/head/localfc/cifp.py","papers":["2106.05519"],"paper_pages":[{"arxiv_id":"2106.05519","page":"/paper/consistent-instance-false-positive-improves"}],"arg_sig_recorded":[["input",2,"float","float32"]],"scalar_args":{"axis":"1"},"class_bearing":false},{"code_sha256_prefix":"27524a6bac90b755","path":"src/model/margin/adaface.py","papers":["2204.00964"],"paper_pages":[{"arxiv_id":"2204.00964","page":"/paper/adaface-quality-adaptive-margin-for-face"}],"arg_sig_recorded":[["input_x",2,"float","float32"]],"scalar_args":{"axis":"1"},"class_bearing":false},{"code_sha256_prefix":"32e701066e89cd01","path":"easyface/recognition/heads/adaface.py","papers":["2204.00964"],"paper_pages":[{"arxiv_id":"2204.00964","page":"/paper/adaface-quality-adaptive-margin-for-face"}],"arg_sig_recorded":[["x",2,"float","float32"]],"scalar_args":{"dim":"1"},"class_bearing":false},{"code_sha256_prefix":"7b15c36502590f85","path":"main_moco.py","papers":["2210.05557"],"paper_pages":[{"arxiv_id":"2210.05557","page":"/paper/opera-omni-supervised-representation-learning"}],"arg_sig_recorded":[["input",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"b3008cda91f30a22","path":"model.py","papers":["1706.07567"],"paper_pages":[{"arxiv_id":"1706.07567","page":"/paper/sampling-matters-in-deep-embedding-learning"}],"arg_sig_recorded":[["x",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"cb8446a797997e25","path":"models/bottom_up_top_down_ranking.py","papers":["1909.04402"],"paper_pages":[{"arxiv_id":"1909.04402","page":"/paper/compositional-generalization-in-image"}],"arg_sig_recorded":[["X",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]},{"output_sha":"52b2d26799446195","size":1,"n_papers":2,"shape":[4],"dtype":"float32","type":"ndarray","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":["norm_value"],"values":[2.9917142391204834,3.138503313064575,2.3749611377716064,2.8242647647857666],"values_recorded":4,"members":[{"code_sha256_prefix":"7f8230b3084eaaab","path":"Normal-Estimation/ZTEE/evaluate.py","papers":["2303.17167","2403.05117"],"paper_pages":[{"arxiv_id":"2303.17167","page":"/paper/rethinking-the-approximation-error-in-3d"},{"arxiv_id":"2403.05117","page":"/paper/arbitrary-scale-point-cloud-upsampling-by"}],"arg_sig_recorded":[["v",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]},{"output_sha":"bd1303ad1c87186e","size":1,"n_papers":2,"shape":[1],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[5.693633556365967],"values_recorded":1,"members":[{"code_sha256_prefix":"bbffec6aadbbfd95","path":"trades.py","papers":["2305.13948","1910.09338"],"paper_pages":[{"arxiv_id":"2305.13948","page":"/paper/decoupled-kullback-leibler-divergence-loss"},{"arxiv_id":"1910.09338","page":"/paper/an-alternative-surrogate-loss-for-pgd-based"}],"arg_sig_recorded":[["x",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]},{"output_sha":"4171f3b5f627ddab","size":1,"n_papers":1,"shape":[4,8],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.31660425662994385,-0.2693844437599182,0.08361149579286575,-0.12134378403425217,-0.1585717648267746,-0.18098656833171844,-0.11640385538339615,0.10061926394701004,0.09113265573978424,0.16845932602882385,-0.06976205110549927,-0.2754564583301544,-0.05067679286003113,0.4078430235385895,-0.09041396528482437,0.09669195860624313,0.011661347001791,-0.28503337502479553,0.07589378952980042,-0.07134035974740982,-0.0039683314971625805,-0.20738644897937775,-0.09429273754358292,0.17276738584041595,-0.13973726332187653,0.07609805464744568,0.16597095131874084,0.17871159315109253,-0.2152905911207199,0.22379697859287262,-0.2322281301021576,-0.10434621572494507],"values_recorded":32,"members":[{"code_sha256_prefix":"a84f15a80e3dbb2a","path":"source/models.py","papers":["2006.08265"],"paper_pages":[{"arxiv_id":"2006.08265","page":"/paper/gs-wgan-a-gradient-sanitized-approach-for"}],"arg_sig_recorded":[["v",2,"float","float32"]],"scalar_args":{"eps":"1e-10"},"class_bearing":false}]}]},{"bucket":[[2,"float","float32"],[2,"float","float32"]],"n_implementations_compared":3,"n_papers":5,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"66687aadf862bd77","size":2,"n_papers":3,"shape":[4],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":2,"torch_reference_conventions_with_this_digest":[],"values":[0.0,0.0,0.0,0.0],"values_recorded":4,"members":[{"code_sha256_prefix":"0fc5d9cde41807d8","path":"model/FME_music_positional_encoding.py","papers":["2311.08355","2212.00973"],"paper_pages":[{"arxiv_id":"2311.08355","page":"/paper/mustango-toward-controllable-text-to-music"},{"arxiv_id":"2212.00973","page":"/paper/a-domain-knowledge-inspired-music-embedding"}],"arg_sig_recorded":[["a",2,"float","float32"],["b",2,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"840f427610121a9d","path":"temos/model/temos.py","papers":["2204.14109"],"paper_pages":[{"arxiv_id":"2204.14109","page":"/paper/temos-generating-diverse-human-motions-from"}],"arg_sig_recorded":[["x1",2,"float","float32"],["x2",2,"float","float32"]],"scalar_args":{"dim":"1"},"class_bearing":false}]},{"output_sha":"fd3f107301960133","size":1,"n_papers":2,"shape":[4,1],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[8.950353622436523,9.850201606750488,5.640440940856934,7.976470947265625],"values_recorded":4,"members":[{"code_sha256_prefix":"e150bca493a582c4","path":"dc_crn.py","papers":["2008.00264","1911.04697"],"paper_pages":[{"arxiv_id":"2008.00264","page":"/paper/dccrn-deep-complex-convolution-recurrent-1"},{"arxiv_id":"1911.04697","page":"/paper/phasen-a-phase-and-harmonics-aware-speech"}],"arg_sig_recorded":[["s1",2,"float","float32"],["s2",2,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]},{"bucket":[[3,"float","float32"]],"n_implementations_compared":3,"n_papers":3,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"da14221e68d4d487","size":2,"n_papers":2,"shape":[2,4,8],"dtype":"float32","type":"ndarray","finite":true,"max_difference_among_recorded_values":1.1920928955078125e-07,"members_with_recorded_values":2,"torch_reference_conventions_with_this_digest":[],"values":[-0.28503453731536865,0.8171771168708801,-0.7414465546607971,-0.2216903418302536,0.10377944260835648,-0.648725688457489,-0.319663941860199,-0.17770570516586304,-0.6827133297920227,0.005765131674706936,-0.500038206577301,-0.6629980206489563,0.5750009417533875,-0.07660806179046631,0.14255593717098236,0.18973077833652496,0.5999174118041992,-0.3619098961353302,-0.051553428173065186,0.6649417877197266,-0.3920564353466034,0.5083616375923157,-0.37922948598861694,0.4461771249771118,0.30456048250198364,0.44856393337249756,-0.4444783926010132,-0.26294389367103577,-0.7105599641799927,0.5611191391944885,-0.8565499186515808,0.8563579320907593,-0.4744567573070526,0.7866631746292114,-0.49397918581962585,-0.5135377049446106,0.4161500334739685,0.44655317068099976,0.0291907899081707,0.7413576245307922,-0.5910013914108276,-0.3779121935367584,-0.4241962730884552,-0.4381965100765228,0.4093991219997406,-0.10187029093503952,0.2583259344100952,0.3849474787712097,-0.16821594536304474,-0.04106158763170242,0.3130861520767212,-0.6374067664146423,0.21993006765842438,0.1334952861070633,-0.4697822034358978,-0.4101235568523407,-0.6303266286849976,0.48647457361221313,0.691389262676239,-0.3714507222175598,0.7815639972686768,0.8788581490516663,-0.843635082244873,-0.3660641312599182],"values_recorded":64,"members":[{"code_sha256_prefix":"2cc7ef3fe9847cad","path":"nlpaug/model/word_embs/word_embeddings.py","papers":["2203.07860"],"paper_pages":[{"arxiv_id":"2203.07860","page":"/paper/imputing-out-of-vocabulary-embeddings-with"}],"arg_sig_recorded":[["data",3,"float","float32"]],"scalar_args":{},"class_bearing":false},{"code_sha256_prefix":"3aa98d060c68358f","path":"Linear-Attention-Mechanism.py","papers":["2007.14902"],"paper_pages":[{"arxiv_id":"2007.14902","page":"/paper/linear-attention-mechanism-an-efficient"}],"arg_sig_recorded":[["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false}]},{"output_sha":"3acc6bb89f3311b5","size":1,"n_papers":1,"shape":[2],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[5.571217060089111,5.7820234298706055],"values_recorded":2,"members":[{"code_sha256_prefix":"6295da24cd9b4a36","path":"5_1_5/test_cifar.py","papers":["2103.02695"],"paper_pages":[{"arxiv_id":"2103.02695","page":"/paper/shift-invariance-can-reduce-adversarial"}],"arg_sig_recorded":[["x",3,"float","float32"]],"scalar_args":{},"class_bearing":false}]}]}]}