{"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/diffaugment","entry":"DiffAugment","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":7,"n_papers":5,"n_buckets":2,"n_distinct_outputs":5,"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/diffaugment","buckets":[{"bucket":[[4,"float","float32"]],"n_implementations_compared":6,"n_papers":5,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"b5326b98d12466cf","size":3,"n_papers":2,"shape":[2,3,4,4],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":0.0,"members_with_recorded_values":3,"torch_reference_conventions_with_this_digest":[],"values":[0.0,0.0,0.5446819067001343,-0.03793797641992569,0.0,0.0,-0.14634643495082855,0.34756043553352356,0.12124542891979218,-0.11997221410274506,-1.0628769397735596,0.22113975882530212,-0.1768953502178192,-0.200820192694664,-0.7058175802230835,-0.0966406837105751,-0.0,0.0,0.4227725565433502,-0.4040393829345703,-0.0,0.0,-0.16435813903808594,0.4335072934627533,0.10414491593837738,0.06820853054523468,-1.1882543563842773,0.19879969954490662,-0.02211352251470089,-0.13793745636940002,-0.19702854752540588,-0.28395432233810425,-0.0,0.0,0.26896318793296814,-0.26425695419311523,-0.0,0.0,-0.07718470692634583,0.05666173994541168,0.04189129173755646,-0.3124347925186157,-0.9988715648651123,0.39689281582832336,-0.07234099507331848,-0.1619063913822174,-0.2774290442466736,-0.39088526368141174,0.0,0.47815200686454773,0.21091784536838531,0.0,0.0,1.5537927150726318,0.2516134977340698,-0.0,0.0,-0.3051825165748596,-0.6939550042152405,0.16553543508052826,0.0,-0.9089005589485168,-0.04727396368980408,-0.2991122603416443,0.0,0.4218740463256836,-0.2581815719604492,0.0,0.0,1.2035088539123535,0.4452497363090515,0.0,0.0,0.29166072607040405,-0.007950454950332642,0.729887843132019,0.0,0.25567716360092163,0.2538631558418274,0.08930180966854095,0.0,-0.4947492480278015,0.04225918650627136,0.0,0.0,1.2730180025100708,0.8070781230926514,-0.0,0.0,0.17856743931770325,0.03845834732055664,1.1661386489868164,0.0,0.07742628455162048,0.21816536784172058,-0.5185861587524414],"values_recorded":96,"members":[{"code_sha256_prefix":"b0bbabd6de30f612","path":"DiffAugment_pytorch.py","papers":["2006.10738","2104.03310"],"paper_pages":[{"arxiv_id":"2006.10738","page":"/paper/differentiable-augmentation-for-data"},{"arxiv_id":"2104.03310","page":"/paper/regularizing-generative-adversarial-networks"}],"arg_sig_recorded":[["x",4,"float","float32"]],"scalar_args":{"policy":"'color,translation,cutout'","channels_first":"True"},"class_bearing":false},{"code_sha256_prefix":"5a98bf7c8a33d10d","path":"DiffAugment_pytorch.py","papers":["2006.10738"],"paper_pages":[{"arxiv_id":"2006.10738","page":"/paper/differentiable-augmentation-for-data"}],"arg_sig_recorded":[["x",4,"float","float32"]],"scalar_args":{"policy":"'color,translation,cutout'","channels_first":"True"},"class_bearing":false},{"code_sha256_prefix":"b13ee69651d4b880","path":"BigGAN and DiffAugGAN/utils/diff_aug.py","papers":["2006.10738"],"paper_pages":[{"arxiv_id":"2006.10738","page":"/paper/differentiable-augmentation-for-data"}],"arg_sig_recorded":[["x",4,"float","float32"]],"scalar_args":{"policy":"'color,translation,cutout'","channels_first":"True"},"class_bearing":false}]},{"output_sha":"0ddf0db7f25680e4","size":1,"n_papers":1,"shape":[2,3,4,4],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.06335625797510147,0.4854263365268707,0.5446819067001343,-0.03793797641992569,0.0,0.6150052547454834,-0.14634643495082855,0.34756043553352356,0.12124542891979218,-0.11997221410274506,-1.0628769397735596,0.22113975882530212,-0.1768953502178192,-0.200820192694664,-0.7058175802230835,-0.0966406837105751,-0.10468719154596329,0.41005176305770874,0.4227725565433502,-0.4040393829345703,-0.0,0.8080849647521973,-0.16435813903808594,0.4335072934627533,0.10414491593837738,0.06820853054523468,-1.1882543563842773,0.19879969954490662,-0.02211352251470089,-0.13793745636940002,-0.19702854752540588,-0.28395432233810425,-0.10016356408596039,0.5379414558410645,0.26896318793296814,-0.26425695419311523,-0.0,0.9268803596496582,-0.07718470692634583,0.05666173994541168,0.04189129173755646,-0.3124347925186157,-0.9988715648651123,0.39689281582832336,-0.07234099507331848,-0.1619063913822174,-0.2774290442466736,-0.39088526368141174,0.0,0.47815200686454773,0.21091784536838531,1.7388732433319092,0.0,1.5537927150726318,0.2516134977340698,-0.686967670917511,0.0,-0.3051825165748596,-0.6939550042152405,0.0,0.0,-0.9089005589485168,-0.04727396368980408,-0.2991122603416443,0.0,0.4218740463256836,-0.2581815719604492,1.340796709060669,0.0,1.2035088539123535,0.4452497363090515,0.9639491438865662,0.0,0.29166072607040405,-0.007950454950332642,0.0,0.0,0.25567716360092163,0.2538631558418274,0.08930180966854095,0.0,-0.4947492480278015,0.04225918650627136,1.159030556678772,0.0,1.2730180025100708,0.8070781230926514,-0.4213244318962097,0.0,0.17856743931770325,0.03845834732055664,0.0,0.0,0.07742628455162048,0.21816536784172058,-0.5185861587524414],"values_recorded":96,"members":[{"code_sha256_prefix":"fb2ae62e229d8e49","path":"pg_modules/discriminator.py","papers":["2111.01007"],"paper_pages":[{"arxiv_id":"2111.01007","page":"/paper/projected-gans-converge-faster"}],"arg_sig_recorded":[["x",4,"float","float32"]],"scalar_args":{"policy":"'color,translation,cutout'","channels_first":"True"},"class_bearing":false}]},{"output_sha":"d8dff36cd4ec80e1","size":1,"n_papers":1,"shape":[2,3,4,4],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.7231047749519348,0.6530793309211731,1.4460861682891846,1.0985400676727295,1.1372957229614258,-0.19992724061012268,-0.2714869976043701,0.8319791555404663,0.3433533310890198,-0.13238386809825897,-1.1905570030212402,-0.012713957577943802,-0.709074079990387,-0.4362029433250427,-2.665717124938965,0.812962532043457,-0.45303258299827576,0.12553228437900543,0.5928415060043335,-1.463806390762329,-0.7250184416770935,1.1514394283294678,-0.3975509703159332,1.4335219860076904,0.22366678714752197,1.184694528579712,-2.0680742263793945,-0.16907231509685516,0.3742450773715973,0.00391392083838582,0.895301342010498,-0.4980470836162567,-0.4213716387748718,1.020633339881897,-0.4836711883544922,-0.4854681193828583,-0.6618722677230835,1.982889175415039,0.2125764787197113,-1.2040232419967651,-0.21204684674739838,-1.479430913925171,-0.7425822019577026,1.2173826694488525,0.022702597081661224,-0.16384494304656982,0.3325777053833008,-1.246457576751709,1.0319796800613403,0.47569993138313293,2.1016592979431152,-0.01877530850470066,1.654322624206543,-0.6244865655899048,-2.408447742462158,-0.21435445547103882,-1.3870023488998413,-2.00654673576355,-1.3687102794647217,-0.026417039334774017,-2.796109437942505,-0.7385192513465881,-0.6348524689674377,0.21286723017692566,0.8440517783164978,-1.0907533168792725,0.7723703980445862,2.0942370891571045,0.4846275746822357,0.022118685767054558,3.1044223308563232,0.0736190527677536,0.6060231924057007,0.28421318531036377,0.5158193111419678,-0.6696537733078003,1.0927388668060303,0.267061322927475,0.6621701121330261,0.2371496558189392,-2.2168080806732178,-0.08749820291996002,0.16540314257144928,-0.8387367129325867,0.7167379260063171,1.2303640842437744,-1.5213911533355713,1.0778614282608032,0.22837327420711517,0.4391850531101227,1.9725817441940308,-1.1686570644378662,0.49750959873199463,0.14785639941692352,-1.3677366971969604,-0.28564757108688354],"values_recorded":96,"members":[{"code_sha256_prefix":"655f5003e174acea","path":"main_run.py","papers":["2605.13475"],"paper_pages":[{"arxiv_id":"2605.13475","page":"/paper/arxiv-2605-13475"}],"arg_sig_recorded":[["x",4,"float","float32"]],"scalar_args":{"param":"<fx_655f5003e174acea__DiffAugment.DiffAugParam object at 0xf"},"class_bearing":false}]},{"output_sha":"ed3922697652cade","size":1,"n_papers":1,"shape":[2,3,4,4],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[-0.026301458477973938,0.2971486747264862,0.2637309730052948,-0.1391400247812271,-0.03891398757696152,0.4780932664871216,-0.08593401312828064,0.17617656290531158,0.05440486595034599,-0.0798100084066391,-0.6831395030021667,0.1642918437719345,-0.06588219106197357,-0.11061930656433105,-0.27048566937446594,-0.1551302969455719,-0.03686210513114929,0.29241177439689636,0.2560696005821228,-0.16214756667613983,-0.055635880678892136,0.4902273416519165,-0.08706595003604889,0.1815778613090515,0.0533301942050457,-0.06798383593559265,-0.69101881980896,0.16288787126541138,-0.05615496635437012,-0.10666745901107788,-0.23851095139980316,-0.16690197587013245,-0.03657781705260277,0.3004489839076996,0.2464035153388977,-0.15336298942565918,-0.055068884044885635,0.4976930022239685,-0.08158756792545319,0.15789510309696198,0.049417879432439804,-0.09190526604652405,-0.6791171431541443,0.17533700168132782,-0.05931149795651436,-0.10817378759384155,-0.2435636967420578,-0.1736220270395279,-0.010539591312408447,-0.12071842700242996,0.9279906749725342,0.39332324266433716,0.868401288986206,0.21520906686782837,-0.21749871969223022,0.3079187870025635,-0.12115604430437088,-0.33254799246788025,0.3319118916988373,-0.44636070728302,-0.32901737093925476,-0.04488537460565567,-0.31783902645111084,0.10815489292144775,-0.014678649604320526,-0.155219167470932,0.8987133502960205,0.43986186385154724,0.8426389694213867,0.22945037484169006,-0.09607905894517899,0.3142613470554352,-0.07726011425256729,-0.2820945382118225,0.3734182119369507,-0.46052783727645874,-0.24336639046669006,-0.02273770421743393,-0.2892724275588989,0.10868971049785614,-0.08209335058927536,-0.1331227421760559,0.8853450417518616,0.3752637803554535,0.8477511405944824,0.25606170296669006,-0.19796150922775269,0.3363795578479767,-0.08557776361703873,-0.27868130803108215,0.4055030643939972,-0.47151827812194824,-0.25647619366645813,-0.02536316215991974,-0.33398061990737915,0.0971752181649208],"values_recorded":96,"members":[{"code_sha256_prefix":"de12e346f3744454","path":"distill.py","papers":["2203.01531"],"paper_pages":[{"arxiv_id":"2203.01531","page":"/paper/cafe-learning-to-condense-dataset-by-aligning"}],"arg_sig_recorded":[["x",4,"float","float32"]],"scalar_args":{"strategy":"'color'","seed":"-1","param":"<fx_de12e346f3744454__DiffAugment.DiffAugParam object at 0xf"},"class_bearing":false}]}]},{"bucket":[[5,"float","float32"]],"n_implementations_compared":1,"n_papers":1,"n_not_digested":0,"not_digested_by_type":{},"clusters":[{"output_sha":"ea226b4adf3b7fcb","size":1,"n_papers":1,"shape":[2,2,3,4,4],"dtype":"float32","type":"Tensor","finite":true,"max_difference_among_recorded_values":null,"members_with_recorded_values":1,"torch_reference_conventions_with_this_digest":[],"values":[0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,-0.0,-0.7877581119537354,0.42003798484802246,0.0,-0.6286313533782959,-0.5719757080078125,0.06214776635169983,0.0,0.8413923382759094,-0.8728291392326355,-0.1920590102672577,0.0,0.0,0.0,0.0,0.0,-0.0,-1.471703052520752,-0.030267149209976196,0.0,-0.02496922016143799,-0.4002639651298523,-0.020396873354911804,0.0,-0.9393311142921448,-0.9769214987754822,0.45788514614105225,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,-0.0,0.8386961817741394,-0.6735373735427856,0.0,-0.28724583983421326,-0.757707953453064,-0.31198686361312866,0.0,0.2917698621749878,0.7719396948814392,0.1948474943637848,0.0,0.0,0.0,0.0,0.0,-0.0,-0.14755073189735413,0.031851500272750854,0.0,-0.047617897391319275,-0.6700558066368103,-0.04078780114650726,0.0,0.21400821208953857,-0.6513611078262329,0.10456162691116333,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.3308810889720917,0.5003777742385864,0.0,0.0,1.0877106189727783,0.09145092964172363,0.0,0.0,0.36738860607147217,0.8868614435195923,0.0,0.0,0.0,0.0,0.0,0.0,0.9398475885391235,0.010012269020080566,0.0,0.0,1.028112530708313,0.5948034524917603,0.0,0.0,0.5972919464111328,0.7731025218963623,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,0.0,-1.0104632377624512,0.3283957242965698,0.0,0.0,0.9831546545028687,0.8257635831832886,0.0,0.0,2.027425527572632,0.22694124281406403,0.0,0.0,0.0,0.0,0.0,0.0,0.02078378200531006,0.7349026203155518,0.0,0.0,-0.3704575002193451,-0.8884382247924805,0.0,0.0,0.32696694135665894,-0.0742289125919342],"values_recorded":192,"members":[{"code_sha256_prefix":"aff3959c75a4b707","path":"diff_augment.py","papers":["2006.10738"],"paper_pages":[{"arxiv_id":"2006.10738","page":"/paper/differentiable-augmentation-for-data"}],"arg_sig_recorded":[["x",5,"float","float32"]],"scalar_args":{"types":"['contrast', 'translation', 'cutout']"},"class_bearing":false}]}]}]}