headless-start's picture
add thesis checkpoints, results and model card
2f6b87a verified
Raw History Blame Contribute Delete
38 kB
{
"best_val_war": 0.9436127744510978,
"best_val_uar": 0.8422065711699508,
"best_val_loss": 0.16450884051903517,
"best_epoch": 28,
"epoch_selection": {
"rule_id": "set_rule_war_then_uar_v1",
"rule": "retain every candidate whose validation WAR is within 1e-4 of the global maximum; within that set retain every candidate whose validation UAR is within 1e-4 of the retained maximum; within that, every candidate whose validation loss is within 1e-4 of the retained minimum; break remaining ties with the declared ordered keys. Order-independent by construction: no pairwise tolerance comparison is made.",
"tolerance": 0.0001,
"tie_breakers": [
"epoch"
],
"n_candidates": 120,
"max_val_war": 0.9436127744510978,
"n_retained_by_war": 1,
"max_val_uar_in_retained": 0.8422065711699508,
"n_retained_by_uar": 1,
"min_val_loss_in_retained": 0.16450884051903517,
"n_retained_by_loss": 1,
"selected": {
"epoch": 28,
"train_loss": 0.1893969083036129,
"val_war": 0.9436127744510978,
"val_uar": 0.8422065711699508,
"val_loss": 0.16450884051903517,
"val_weighted_f1": 0.9431165336357188,
"updates": 552
}
},
"epochs_completed": 120,
"n_retained_max": 2,
"retained_epochs": [
28
],
"params": {
"total": 85803270,
"trainable": 85803270,
"trainable_pct": 100.0
},
"param_groups": {
"decay": {
"tensors": 50,
"params": 85529088
},
"no_decay": {
"tensors": 102,
"params": 274182
},
"weight_decay": 0.05,
"rule": "architecture no_weight_decay() list, plus every tensor of rank <= 1",
"model_skip_list": [
"cls_token",
"dist_token",
"pos_embed"
],
"frozen_tensors": 0,
"no_decay_names": [
"cls_token",
"pos_embed",
"patch_embed.proj.bias",
"blocks.0.norm1.weight",
"blocks.0.norm1.bias",
"blocks.0.attn.qkv.bias",
"blocks.0.attn.proj.bias",
"blocks.0.norm2.weight",
"blocks.0.norm2.bias",
"blocks.0.mlp.fc1.bias",
"blocks.0.mlp.fc2.bias",
"blocks.1.norm1.weight",
"blocks.1.norm1.bias",
"blocks.1.attn.qkv.bias",
"blocks.1.attn.proj.bias",
"blocks.1.norm2.weight",
"blocks.1.norm2.bias",
"blocks.1.mlp.fc1.bias",
"blocks.1.mlp.fc2.bias",
"blocks.2.norm1.weight",
"blocks.2.norm1.bias",
"blocks.2.attn.qkv.bias",
"blocks.2.attn.proj.bias",
"blocks.2.norm2.weight",
"blocks.2.norm2.bias",
"blocks.2.mlp.fc1.bias",
"blocks.2.mlp.fc2.bias",
"blocks.3.norm1.weight",
"blocks.3.norm1.bias",
"blocks.3.attn.qkv.bias",
"blocks.3.attn.proj.bias",
"blocks.3.norm2.weight",
"blocks.3.norm2.bias",
"blocks.3.mlp.fc1.bias",
"blocks.3.mlp.fc2.bias",
"blocks.4.norm1.weight",
"blocks.4.norm1.bias",
"blocks.4.attn.qkv.bias",
"blocks.4.attn.proj.bias",
"blocks.4.norm2.weight",
"blocks.4.norm2.bias",
"blocks.4.mlp.fc1.bias",
"blocks.4.mlp.fc2.bias",
"blocks.5.norm1.weight",
"blocks.5.norm1.bias",
"blocks.5.attn.qkv.bias",
"blocks.5.attn.proj.bias",
"blocks.5.norm2.weight",
"blocks.5.norm2.bias",
"blocks.5.mlp.fc1.bias",
"blocks.5.mlp.fc2.bias",
"blocks.6.norm1.weight",
"blocks.6.norm1.bias",
"blocks.6.attn.qkv.bias",
"blocks.6.attn.proj.bias",
"blocks.6.norm2.weight",
"blocks.6.norm2.bias",
"blocks.6.mlp.fc1.bias",
"blocks.6.mlp.fc2.bias",
"blocks.7.norm1.weight",
"blocks.7.norm1.bias",
"blocks.7.attn.qkv.bias",
"blocks.7.attn.proj.bias",
"blocks.7.norm2.weight",
"blocks.7.norm2.bias",
"blocks.7.mlp.fc1.bias",
"blocks.7.mlp.fc2.bias",
"blocks.8.norm1.weight",
"blocks.8.norm1.bias",
"blocks.8.attn.qkv.bias",
"blocks.8.attn.proj.bias",
"blocks.8.norm2.weight",
"blocks.8.norm2.bias",
"blocks.8.mlp.fc1.bias",
"blocks.8.mlp.fc2.bias",
"blocks.9.norm1.weight",
"blocks.9.norm1.bias",
"blocks.9.attn.qkv.bias",
"blocks.9.attn.proj.bias",
"blocks.9.norm2.weight",
"blocks.9.norm2.bias",
"blocks.9.mlp.fc1.bias",
"blocks.9.mlp.fc2.bias",
"blocks.10.norm1.weight",
"blocks.10.norm1.bias",
"blocks.10.attn.qkv.bias",
"blocks.10.attn.proj.bias",
"blocks.10.norm2.weight",
"blocks.10.norm2.bias",
"blocks.10.mlp.fc1.bias",
"blocks.10.mlp.fc2.bias",
"blocks.11.norm1.weight",
"blocks.11.norm1.bias",
"blocks.11.attn.qkv.bias",
"blocks.11.attn.proj.bias",
"blocks.11.norm2.weight",
"blocks.11.norm2.bias",
"blocks.11.mlp.fc1.bias",
"blocks.11.mlp.fc2.bias",
"norm.weight",
"norm.bias",
"head.bias"
],
"total_steps": 66240,
"warmup_steps": 1104
},
"run_identity": {
"method": "full_ft",
"protocol": "official_split",
"split_seed": 42,
"model_seed": 42,
"source_stage": "augreg",
"init_from_sha256": null,
"data_root": "data/ferplus"
},
"checkpoint_sha256": "3094f103c17bd13558f960c22b91ed3316679cadeab88ba269d862019c4dd58a",
"amp_active": true,
"train_time_sec": 18801.249953985214,
"time_per_epoch_sec": 156.6770829498768,
"peak_mem_gb": 3.2119665145874023,
"skipped_steps": 28,
"optimizer_updates": 66212,
"history": [
{
"epoch": 1,
"train_loss": 1.4047704333016626,
"val_war": 0.8223552894211577,
"val_uar": 0.6159856467505688,
"val_loss": 0.5097239920479095,
"val_weighted_f1": 0.8197161181996515,
"updates": 547
},
{
"epoch": 2,
"train_loss": 0.6416764952992455,
"val_war": 0.8837325349301397,
"val_uar": 0.651311836222769,
"val_loss": 0.3376162086061375,
"val_weighted_f1": 0.8729204991746009,
"updates": 552
},
{
"epoch": 3,
"train_loss": 0.4902889822945771,
"val_war": 0.9106786427145709,
"val_uar": 0.7028063560588943,
"val_loss": 0.2467307984650492,
"val_weighted_f1": 0.9039909599003104,
"updates": 552
},
{
"epoch": 4,
"train_loss": 0.41320013329804595,
"val_war": 0.9101796407185628,
"val_uar": 0.758151098367541,
"val_loss": 0.23912653208016874,
"val_weighted_f1": 0.9078919964997306,
"updates": 552
},
{
"epoch": 5,
"train_loss": 0.3782459644850952,
"val_war": 0.9276447105788423,
"val_uar": 0.7527450839795714,
"val_loss": 0.19767410455349677,
"val_weighted_f1": 0.9233663934684276,
"updates": 552
},
{
"epoch": 6,
"train_loss": 0.3450085104919954,
"val_war": 0.9316367265469062,
"val_uar": 0.774065657914138,
"val_loss": 0.19791446635109222,
"val_weighted_f1": 0.9285206478898584,
"updates": 552
},
{
"epoch": 7,
"train_loss": 0.3270069262146119,
"val_war": 0.9336327345309381,
"val_uar": 0.7944838654815302,
"val_loss": 0.1838007161687472,
"val_weighted_f1": 0.9319617105616057,
"updates": 552
},
{
"epoch": 8,
"train_loss": 0.3104032961829284,
"val_war": 0.9266467065868264,
"val_uar": 0.7657295042459896,
"val_loss": 0.19788938168874995,
"val_weighted_f1": 0.9232278845048117,
"updates": 552
},
{
"epoch": 9,
"train_loss": 0.29067675521806524,
"val_war": 0.939620758483034,
"val_uar": 0.8348326702929394,
"val_loss": 0.18219371157848907,
"val_weighted_f1": 0.9392222992123364,
"updates": 552
},
{
"epoch": 10,
"train_loss": 0.28042608418145076,
"val_war": 0.9336327345309381,
"val_uar": 0.82131829483203,
"val_loss": 0.177239563234314,
"val_weighted_f1": 0.931834378555007,
"updates": 552
},
{
"epoch": 11,
"train_loss": 0.262686960672253,
"val_war": 0.939620758483034,
"val_uar": 0.8038431213620192,
"val_loss": 0.17341571445474604,
"val_weighted_f1": 0.9376495320877638,
"updates": 552
},
{
"epoch": 12,
"train_loss": 0.26096923651759274,
"val_war": 0.9356287425149701,
"val_uar": 0.8421289800253589,
"val_loss": 0.1788175468316335,
"val_weighted_f1": 0.9353055837721055,
"updates": 552
},
{
"epoch": 13,
"train_loss": 0.25120533699118364,
"val_war": 0.9391217564870259,
"val_uar": 0.8051524911546393,
"val_loss": 0.1654306911601278,
"val_weighted_f1": 0.9375012220426306,
"updates": 552
},
{
"epoch": 14,
"train_loss": 0.24476321863254186,
"val_war": 0.9406187624750499,
"val_uar": 0.8326409383075886,
"val_loss": 0.167763271701788,
"val_weighted_f1": 0.9392802550828819,
"updates": 552
},
{
"epoch": 15,
"train_loss": 0.237078241127205,
"val_war": 0.936127744510978,
"val_uar": 0.8122128463652457,
"val_loss": 0.17576297711647912,
"val_weighted_f1": 0.9343963572417492,
"updates": 551
},
{
"epoch": 16,
"train_loss": 0.22895441080083875,
"val_war": 0.937624750499002,
"val_uar": 0.8150744750602922,
"val_loss": 0.17539126556850004,
"val_weighted_f1": 0.9359240875153829,
"updates": 552
},
{
"epoch": 17,
"train_loss": 0.22731330655302032,
"val_war": 0.9346307385229541,
"val_uar": 0.8057130695171434,
"val_loss": 0.17710513801512842,
"val_weighted_f1": 0.9327008395248881,
"updates": 552
},
{
"epoch": 18,
"train_loss": 0.21861273906168927,
"val_war": 0.939620758483034,
"val_uar": 0.8428411018338874,
"val_loss": 0.16258980455750716,
"val_weighted_f1": 0.9387422093235203,
"updates": 552
},
{
"epoch": 19,
"train_loss": 0.2129864644221858,
"val_war": 0.9406187624750499,
"val_uar": 0.8102733595939228,
"val_loss": 0.16628249388136074,
"val_weighted_f1": 0.9389083202499009,
"updates": 551
},
{
"epoch": 20,
"train_loss": 0.21217662969697104,
"val_war": 0.938622754491018,
"val_uar": 0.8278308407678868,
"val_loss": 0.16899946621970502,
"val_weighted_f1": 0.9374319633540159,
"updates": 552
},
{
"epoch": 21,
"train_loss": 0.20610570450874843,
"val_war": 0.9341317365269461,
"val_uar": 0.8014836738025538,
"val_loss": 0.17031347205836378,
"val_weighted_f1": 0.9324509759191477,
"updates": 552
},
{
"epoch": 22,
"train_loss": 0.2021111387766412,
"val_war": 0.937125748502994,
"val_uar": 0.8233250264680203,
"val_loss": 0.17300484442484831,
"val_weighted_f1": 0.9356376442336026,
"updates": 552
},
{
"epoch": 23,
"train_loss": 0.20047183184811201,
"val_war": 0.9416167664670658,
"val_uar": 0.8327062078589332,
"val_loss": 0.1606178121444232,
"val_weighted_f1": 0.9403728956253352,
"updates": 551
},
{
"epoch": 24,
"train_loss": 0.19719777391957236,
"val_war": 0.9421157684630739,
"val_uar": 0.842356635721179,
"val_loss": 0.1603443944852747,
"val_weighted_f1": 0.9416712024861985,
"updates": 552
},
{
"epoch": 25,
"train_loss": 0.19344528113726578,
"val_war": 0.935129740518962,
"val_uar": 0.8543851100798032,
"val_loss": 0.17777673513470535,
"val_weighted_f1": 0.9348537743649601,
"updates": 552
},
{
"epoch": 26,
"train_loss": 0.1901420341960486,
"val_war": 0.9331337325349301,
"val_uar": 0.7908487120788793,
"val_loss": 0.17135679430352477,
"val_weighted_f1": 0.930950739360817,
"updates": 552
},
{
"epoch": 27,
"train_loss": 0.19049885533790156,
"val_war": 0.938622754491018,
"val_uar": 0.8379346014513839,
"val_loss": 0.1668140509736514,
"val_weighted_f1": 0.9383352279026699,
"updates": 551
},
{
"epoch": 28,
"train_loss": 0.1893969083036129,
"val_war": 0.9436127744510978,
"val_uar": 0.8422065711699508,
"val_loss": 0.16450884051903517,
"val_weighted_f1": 0.9431165336357188,
"updates": 552
},
{
"epoch": 29,
"train_loss": 0.1831858852728049,
"val_war": 0.9311377245508982,
"val_uar": 0.8308813161602101,
"val_loss": 0.19041404312480234,
"val_weighted_f1": 0.9299362619686248,
"updates": 552
},
{
"epoch": 30,
"train_loss": 0.18308283137377582,
"val_war": 0.939620758483034,
"val_uar": 0.8314178853183338,
"val_loss": 0.16331172422199192,
"val_weighted_f1": 0.93878645372026,
"updates": 552
},
{
"epoch": 31,
"train_loss": 0.18197595729050398,
"val_war": 0.9321357285429142,
"val_uar": 0.7967183493197861,
"val_loss": 0.17413552166399604,
"val_weighted_f1": 0.929998871426466,
"updates": 552
},
{
"epoch": 32,
"train_loss": 0.17810741992778437,
"val_war": 0.9356287425149701,
"val_uar": 0.807824984251555,
"val_loss": 0.1836102489166393,
"val_weighted_f1": 0.9332500180606241,
"updates": 552
},
{
"epoch": 33,
"train_loss": 0.17732757957780215,
"val_war": 0.938622754491018,
"val_uar": 0.8359269027357614,
"val_loss": 0.1672114410145792,
"val_weighted_f1": 0.9373836502541006,
"updates": 552
},
{
"epoch": 34,
"train_loss": 0.17661268946852782,
"val_war": 0.9346307385229541,
"val_uar": 0.8154203544688675,
"val_loss": 0.1725583569910712,
"val_weighted_f1": 0.93333122447792,
"updates": 551
},
{
"epoch": 35,
"train_loss": 0.17519370785877797,
"val_war": 0.937125748502994,
"val_uar": 0.8297162145398028,
"val_loss": 0.17537833521228113,
"val_weighted_f1": 0.9363643266716352,
"updates": 552
},
{
"epoch": 36,
"train_loss": 0.1711410973390961,
"val_war": 0.936127744510978,
"val_uar": 0.8217137410289959,
"val_loss": 0.17358410074384864,
"val_weighted_f1": 0.9349008181696952,
"updates": 552
},
{
"epoch": 37,
"train_loss": 0.17219136656428133,
"val_war": 0.9411177644710579,
"val_uar": 0.8270144419891348,
"val_loss": 0.16602112324592358,
"val_weighted_f1": 0.9395836841972692,
"updates": 552
},
{
"epoch": 38,
"train_loss": 0.16979345952638533,
"val_war": 0.9236526946107785,
"val_uar": 0.8008375266113957,
"val_loss": 0.18842854220680372,
"val_weighted_f1": 0.9227581205840718,
"updates": 550
},
{
"epoch": 39,
"train_loss": 0.16958456639603348,
"val_war": 0.9346307385229541,
"val_uar": 0.8113874086070426,
"val_loss": 0.18208533474904096,
"val_weighted_f1": 0.9329059251789252,
"updates": 552
},
{
"epoch": 40,
"train_loss": 0.1659080766187424,
"val_war": 0.9331337325349301,
"val_uar": 0.8158811814418448,
"val_loss": 0.18062037025799532,
"val_weighted_f1": 0.9316418452003805,
"updates": 552
},
{
"epoch": 41,
"train_loss": 0.16519805883967528,
"val_war": 0.937624750499002,
"val_uar": 0.8225274975838978,
"val_loss": 0.17126444376216915,
"val_weighted_f1": 0.936323868589868,
"updates": 552
},
{
"epoch": 42,
"train_loss": 0.1648560220337773,
"val_war": 0.9366267465069861,
"val_uar": 0.8340993713289312,
"val_loss": 0.1728722200749401,
"val_weighted_f1": 0.9355585459186554,
"updates": 552
},
{
"epoch": 43,
"train_loss": 0.16293427874502206,
"val_war": 0.9311377245508982,
"val_uar": 0.8229728037252396,
"val_loss": 0.18072660178124547,
"val_weighted_f1": 0.9299140972565592,
"updates": 552
},
{
"epoch": 44,
"train_loss": 0.16389999659551008,
"val_war": 0.9331337325349301,
"val_uar": 0.8378471805786791,
"val_loss": 0.17224340731214857,
"val_weighted_f1": 0.9322058935057838,
"updates": 552
},
{
"epoch": 45,
"train_loss": 0.161916443627807,
"val_war": 0.935129740518962,
"val_uar": 0.8314001098910526,
"val_loss": 0.17295021951912407,
"val_weighted_f1": 0.934366217306364,
"updates": 551
},
{
"epoch": 46,
"train_loss": 0.16076805402007321,
"val_war": 0.9291417165668663,
"val_uar": 0.8174550410960005,
"val_loss": 0.18424697141506952,
"val_weighted_f1": 0.9273992850694578,
"updates": 552
},
{
"epoch": 47,
"train_loss": 0.15887886772722531,
"val_war": 0.9346307385229541,
"val_uar": 0.8365652156390958,
"val_loss": 0.1765581264705239,
"val_weighted_f1": 0.9334462961038144,
"updates": 552
},
{
"epoch": 48,
"train_loss": 0.15737028310519563,
"val_war": 0.9336327345309381,
"val_uar": 0.8169661806681837,
"val_loss": 0.17476384503815226,
"val_weighted_f1": 0.9322262324323962,
"updates": 552
},
{
"epoch": 49,
"train_loss": 0.15786523003479178,
"val_war": 0.9406187624750499,
"val_uar": 0.8254044970943709,
"val_loss": 0.1667531690971104,
"val_weighted_f1": 0.939319809871151,
"updates": 551
},
{
"epoch": 50,
"train_loss": 0.1579624950213781,
"val_war": 0.936127744510978,
"val_uar": 0.8327343408522303,
"val_loss": 0.17719584872979605,
"val_weighted_f1": 0.9346604369692918,
"updates": 552
},
{
"epoch": 51,
"train_loss": 0.15547056047424304,
"val_war": 0.9426147704590818,
"val_uar": 0.8433774102687055,
"val_loss": 0.16802481837020425,
"val_weighted_f1": 0.9420444338655707,
"updates": 552
},
{
"epoch": 52,
"train_loss": 0.15682258752461337,
"val_war": 0.936127744510978,
"val_uar": 0.8350100725401344,
"val_loss": 0.16921452003086873,
"val_weighted_f1": 0.9350474432847485,
"updates": 552
},
{
"epoch": 53,
"train_loss": 0.15364923303130396,
"val_war": 0.9311377245508982,
"val_uar": 0.8073020137932175,
"val_loss": 0.18136151143890655,
"val_weighted_f1": 0.9296318489900549,
"updates": 551
},
{
"epoch": 54,
"train_loss": 0.15273373563254794,
"val_war": 0.9406187624750499,
"val_uar": 0.8307596572682553,
"val_loss": 0.16410571304029095,
"val_weighted_f1": 0.939278460281594,
"updates": 552
},
{
"epoch": 55,
"train_loss": 0.1523624458734621,
"val_war": 0.9331337325349301,
"val_uar": 0.8336914121804714,
"val_loss": 0.18340963726034185,
"val_weighted_f1": 0.9323174517499838,
"updates": 552
},
{
"epoch": 56,
"train_loss": 0.1521285242463271,
"val_war": 0.936127744510978,
"val_uar": 0.8597611370910458,
"val_loss": 0.171738973649557,
"val_weighted_f1": 0.9358443761140609,
"updates": 552
},
{
"epoch": 57,
"train_loss": 0.149947994185449,
"val_war": 0.937624750499002,
"val_uar": 0.8275117786242937,
"val_loss": 0.17215411481029258,
"val_weighted_f1": 0.9366187640703773,
"updates": 551
},
{
"epoch": 58,
"train_loss": 0.14973477020451964,
"val_war": 0.9356287425149701,
"val_uar": 0.8265770511627251,
"val_loss": 0.17521684984128155,
"val_weighted_f1": 0.9348368439643246,
"updates": 552
},
{
"epoch": 59,
"train_loss": 0.14835869101203336,
"val_war": 0.939620758483034,
"val_uar": 0.8121690458857072,
"val_loss": 0.17793412170486297,
"val_weighted_f1": 0.9377800098595549,
"updates": 552
},
{
"epoch": 60,
"train_loss": 0.14778099850205942,
"val_war": 0.937125748502994,
"val_uar": 0.8374458697734122,
"val_loss": 0.18030216605839378,
"val_weighted_f1": 0.9359710989521886,
"updates": 551
},
{
"epoch": 61,
"train_loss": 0.14852110827791745,
"val_war": 0.9366267465069861,
"val_uar": 0.8326849490326813,
"val_loss": 0.1811545911782278,
"val_weighted_f1": 0.9354929700443406,
"updates": 552
},
{
"epoch": 62,
"train_loss": 0.14789182282606472,
"val_war": 0.9411177644710579,
"val_uar": 0.8379392000499566,
"val_loss": 0.18187752142994704,
"val_weighted_f1": 0.9399631953367963,
"updates": 552
},
{
"epoch": 63,
"train_loss": 0.1469726145730114,
"val_war": 0.9321357285429142,
"val_uar": 0.8046150840463109,
"val_loss": 0.19261429753310666,
"val_weighted_f1": 0.930283583662995,
"updates": 552
},
{
"epoch": 64,
"train_loss": 0.14675609817076565,
"val_war": 0.93812375249501,
"val_uar": 0.8181238952943793,
"val_loss": 0.1784201398224174,
"val_weighted_f1": 0.9363240961445806,
"updates": 552
},
{
"epoch": 65,
"train_loss": 0.14619851021528663,
"val_war": 0.9401197604790419,
"val_uar": 0.8114451914341151,
"val_loss": 0.17476017050400466,
"val_weighted_f1": 0.9383031809394751,
"updates": 552
},
{
"epoch": 66,
"train_loss": 0.14603796090125956,
"val_war": 0.9366267465069861,
"val_uar": 0.8074483129594299,
"val_loss": 0.18011196317549,
"val_weighted_f1": 0.9348054209652275,
"updates": 552
},
{
"epoch": 67,
"train_loss": 0.14329448090157598,
"val_war": 0.9356287425149701,
"val_uar": 0.8106711128985123,
"val_loss": 0.1846027246968237,
"val_weighted_f1": 0.9337210658403128,
"updates": 552
},
{
"epoch": 68,
"train_loss": 0.1428006283184002,
"val_war": 0.9316367265469062,
"val_uar": 0.8031366712868886,
"val_loss": 0.18349386024439407,
"val_weighted_f1": 0.9300153443926755,
"updates": 551
},
{
"epoch": 69,
"train_loss": 0.14301485549754567,
"val_war": 0.9391217564870259,
"val_uar": 0.8263765855244181,
"val_loss": 0.17007501188390506,
"val_weighted_f1": 0.9379734352383235,
"updates": 552
},
{
"epoch": 70,
"train_loss": 0.1432912688790776,
"val_war": 0.9346307385229541,
"val_uar": 0.8111892343577506,
"val_loss": 0.18101948895259293,
"val_weighted_f1": 0.9330041838798201,
"updates": 552
},
{
"epoch": 71,
"train_loss": 0.14144405708899682,
"val_war": 0.93812375249501,
"val_uar": 0.8278699721986621,
"val_loss": 0.17523026980920703,
"val_weighted_f1": 0.9371255511915308,
"updates": 552
},
{
"epoch": 72,
"train_loss": 0.1411079731306863,
"val_war": 0.936127744510978,
"val_uar": 0.8390960814625954,
"val_loss": 0.17221887738345387,
"val_weighted_f1": 0.9357038867595254,
"updates": 551
},
{
"epoch": 73,
"train_loss": 0.141164897639828,
"val_war": 0.937624750499002,
"val_uar": 0.8268934771477933,
"val_loss": 0.17504457956897523,
"val_weighted_f1": 0.9361964607187647,
"updates": 552
},
{
"epoch": 74,
"train_loss": 0.1412509002880931,
"val_war": 0.937624750499002,
"val_uar": 0.8344505680864871,
"val_loss": 0.173524490224863,
"val_weighted_f1": 0.936921531468762,
"updates": 552
},
{
"epoch": 75,
"train_loss": 0.13983115238121133,
"val_war": 0.9346307385229541,
"val_uar": 0.8339003855987236,
"val_loss": 0.179867926739647,
"val_weighted_f1": 0.9336630157149802,
"updates": 552
},
{
"epoch": 76,
"train_loss": 0.14020412397073298,
"val_war": 0.937624750499002,
"val_uar": 0.8262726983698517,
"val_loss": 0.17286987647324978,
"val_weighted_f1": 0.9364524259053449,
"updates": 551
},
{
"epoch": 77,
"train_loss": 0.13932993502339913,
"val_war": 0.939620758483034,
"val_uar": 0.8414604555357621,
"val_loss": 0.17529048795947533,
"val_weighted_f1": 0.9386938672967305,
"updates": 552
},
{
"epoch": 78,
"train_loss": 0.13933016522991967,
"val_war": 0.937125748502994,
"val_uar": 0.8275055800868318,
"val_loss": 0.1820735025429678,
"val_weighted_f1": 0.9356624613816862,
"updates": 552
},
{
"epoch": 79,
"train_loss": 0.13874173161328518,
"val_war": 0.9411177644710579,
"val_uar": 0.830963406519445,
"val_loss": 0.1751293259972108,
"val_weighted_f1": 0.9400655519928833,
"updates": 551
},
{
"epoch": 80,
"train_loss": 0.13859787873970633,
"val_war": 0.9391217564870259,
"val_uar": 0.8451352078232186,
"val_loss": 0.17977165573549841,
"val_weighted_f1": 0.9384386366593882,
"updates": 552
},
{
"epoch": 81,
"train_loss": 0.13631444165598036,
"val_war": 0.939620758483034,
"val_uar": 0.832240165663004,
"val_loss": 0.1811827164061531,
"val_weighted_f1": 0.938325838303113,
"updates": 552
},
{
"epoch": 82,
"train_loss": 0.13703395100576898,
"val_war": 0.9411177644710579,
"val_uar": 0.8304851210394378,
"val_loss": 0.1771898951656566,
"val_weighted_f1": 0.9401466011116297,
"updates": 552
},
{
"epoch": 83,
"train_loss": 0.13599931567773949,
"val_war": 0.936127744510978,
"val_uar": 0.8188222957698937,
"val_loss": 0.18242532733136307,
"val_weighted_f1": 0.9346576422860918,
"updates": 551
},
{
"epoch": 84,
"train_loss": 0.13594216169147516,
"val_war": 0.939620758483034,
"val_uar": 0.8317135786104174,
"val_loss": 0.1794372118132081,
"val_weighted_f1": 0.9381984900340161,
"updates": 552
},
{
"epoch": 85,
"train_loss": 0.13535929553944775,
"val_war": 0.9406187624750499,
"val_uar": 0.8263293364063524,
"val_loss": 0.17809502831000293,
"val_weighted_f1": 0.9391574440528745,
"updates": 552
},
{
"epoch": 86,
"train_loss": 0.13495322340902785,
"val_war": 0.9366267465069861,
"val_uar": 0.8284884049407508,
"val_loss": 0.18140700646860156,
"val_weighted_f1": 0.9353187792505459,
"updates": 552
},
{
"epoch": 87,
"train_loss": 0.1354656302687818,
"val_war": 0.9391217564870259,
"val_uar": 0.8247989486971029,
"val_loss": 0.1832676084931501,
"val_weighted_f1": 0.937619021017889,
"updates": 552
},
{
"epoch": 88,
"train_loss": 0.13511739565197464,
"val_war": 0.9391217564870259,
"val_uar": 0.8168166353390379,
"val_loss": 0.1806866105593845,
"val_weighted_f1": 0.937421845224609,
"updates": 551
},
{
"epoch": 89,
"train_loss": 0.13420768714540032,
"val_war": 0.937624750499002,
"val_uar": 0.8182509202713549,
"val_loss": 0.17914717357434676,
"val_weighted_f1": 0.9362079011897461,
"updates": 552
},
{
"epoch": 90,
"train_loss": 0.13419607986263038,
"val_war": 0.939620758483034,
"val_uar": 0.820215573308591,
"val_loss": 0.18154250928563273,
"val_weighted_f1": 0.9380655094948194,
"updates": 552
},
{
"epoch": 91,
"train_loss": 0.133312442309829,
"val_war": 0.937624750499002,
"val_uar": 0.8252876864403995,
"val_loss": 0.182653937778787,
"val_weighted_f1": 0.93646878054044,
"updates": 552
},
{
"epoch": 92,
"train_loss": 0.13295968446162368,
"val_war": 0.938622754491018,
"val_uar": 0.8235300876715872,
"val_loss": 0.17950444788989905,
"val_weighted_f1": 0.9372398945588261,
"updates": 552
},
{
"epoch": 93,
"train_loss": 0.13368176660735737,
"val_war": 0.935129740518962,
"val_uar": 0.8109127337591948,
"val_loss": 0.18271107064511724,
"val_weighted_f1": 0.9332663929251748,
"updates": 552
},
{
"epoch": 94,
"train_loss": 0.13279324432373937,
"val_war": 0.9356287425149701,
"val_uar": 0.8206186113891706,
"val_loss": 0.17936357733851183,
"val_weighted_f1": 0.9343506117826228,
"updates": 552
},
{
"epoch": 95,
"train_loss": 0.13153805357087106,
"val_war": 0.938622754491018,
"val_uar": 0.8063027351163772,
"val_loss": 0.18261967840308915,
"val_weighted_f1": 0.9366194779986263,
"updates": 551
},
{
"epoch": 96,
"train_loss": 0.13274440862684572,
"val_war": 0.9341317365269461,
"val_uar": 0.8105302216756279,
"val_loss": 0.18330342991801793,
"val_weighted_f1": 0.932406482401329,
"updates": 552
},
{
"epoch": 97,
"train_loss": 0.1322247357060265,
"val_war": 0.93812375249501,
"val_uar": 0.8216714461193625,
"val_loss": 0.18368515761907467,
"val_weighted_f1": 0.936473096566982,
"updates": 552
},
{
"epoch": 98,
"train_loss": 0.13247279008922952,
"val_war": 0.9401197604790419,
"val_uar": 0.8232914701423778,
"val_loss": 0.1798601560815128,
"val_weighted_f1": 0.938497165276114,
"updates": 552
},
{
"epoch": 99,
"train_loss": 0.1319061137681392,
"val_war": 0.938622754491018,
"val_uar": 0.8274636360120335,
"val_loss": 0.17892851613595814,
"val_weighted_f1": 0.9371650037471614,
"updates": 551
},
{
"epoch": 100,
"train_loss": 0.13156906532623985,
"val_war": 0.939620758483034,
"val_uar": 0.8294923307976768,
"val_loss": 0.18129546410904196,
"val_weighted_f1": 0.9381280097928196,
"updates": 552
},
{
"epoch": 101,
"train_loss": 0.13197330744784788,
"val_war": 0.938622754491018,
"val_uar": 0.8283308906118463,
"val_loss": 0.1829780256617331,
"val_weighted_f1": 0.937134316978444,
"updates": 552
},
{
"epoch": 102,
"train_loss": 0.13150832892218958,
"val_war": 0.9401197604790419,
"val_uar": 0.8376741282212503,
"val_loss": 0.18057064132062262,
"val_weighted_f1": 0.9388598979782949,
"updates": 552
},
{
"epoch": 103,
"train_loss": 0.13130041319184868,
"val_war": 0.9416167664670658,
"val_uar": 0.8326645222829119,
"val_loss": 0.17874055357274418,
"val_weighted_f1": 0.9401820757548326,
"updates": 552
},
{
"epoch": 104,
"train_loss": 0.1309262527849686,
"val_war": 0.9391217564870259,
"val_uar": 0.8192555209490798,
"val_loss": 0.1813111626459453,
"val_weighted_f1": 0.9374083180236855,
"updates": 551
},
{
"epoch": 105,
"train_loss": 0.13059712380540586,
"val_war": 0.939620758483034,
"val_uar": 0.8355117432097643,
"val_loss": 0.18196073498673543,
"val_weighted_f1": 0.9384874870992324,
"updates": 552
},
{
"epoch": 106,
"train_loss": 0.1309271764687037,
"val_war": 0.939620758483034,
"val_uar": 0.8309386119751484,
"val_loss": 0.1815571985439864,
"val_weighted_f1": 0.9382253811515258,
"updates": 552
},
{
"epoch": 107,
"train_loss": 0.13030955485033385,
"val_war": 0.937624750499002,
"val_uar": 0.8375510215242851,
"val_loss": 0.18260493749629952,
"val_weighted_f1": 0.9364580918966305,
"updates": 552
},
{
"epoch": 108,
"train_loss": 0.13109447989802978,
"val_war": 0.938622754491018,
"val_uar": 0.8295682435909755,
"val_loss": 0.18228361816582328,
"val_weighted_f1": 0.9371732495935052,
"updates": 552
},
{
"epoch": 109,
"train_loss": 0.1309809476484554,
"val_war": 0.93812375249501,
"val_uar": 0.8295546790308276,
"val_loss": 0.1809846424979126,
"val_weighted_f1": 0.936760537702893,
"updates": 551
},
{
"epoch": 110,
"train_loss": 0.13025828525673044,
"val_war": 0.9366267465069861,
"val_uar": 0.8212696807333403,
"val_loss": 0.18340231420215733,
"val_weighted_f1": 0.9350511961723915,
"updates": 552
},
{
"epoch": 111,
"train_loss": 0.1309905729330597,
"val_war": 0.9391217564870259,
"val_uar": 0.8305370055494455,
"val_loss": 0.18326309073471023,
"val_weighted_f1": 0.9377237998088513,
"updates": 552
},
{
"epoch": 112,
"train_loss": 0.13059631766364443,
"val_war": 0.9401197604790419,
"val_uar": 0.8289526552200405,
"val_loss": 0.18176157885563826,
"val_weighted_f1": 0.9387643757048428,
"updates": 552
},
{
"epoch": 113,
"train_loss": 0.13075721938789658,
"val_war": 0.937624750499002,
"val_uar": 0.8224311209191706,
"val_loss": 0.1838092332977259,
"val_weighted_f1": 0.936078411773883,
"updates": 552
},
{
"epoch": 114,
"train_loss": 0.13033562467581936,
"val_war": 0.9391217564870259,
"val_uar": 0.8283881850981074,
"val_loss": 0.1832540118825174,
"val_weighted_f1": 0.9377460738926298,
"updates": 552
},
{
"epoch": 115,
"train_loss": 0.13043150780743076,
"val_war": 0.938622754491018,
"val_uar": 0.8254178574742898,
"val_loss": 0.18307361018633891,
"val_weighted_f1": 0.9371398367088126,
"updates": 552
},
{
"epoch": 116,
"train_loss": 0.1303248384266857,
"val_war": 0.93812375249501,
"val_uar": 0.8275117448642374,
"val_loss": 0.18226855175342865,
"val_weighted_f1": 0.9367804571542246,
"updates": 551
},
{
"epoch": 117,
"train_loss": 0.13041031554794658,
"val_war": 0.9391217564870259,
"val_uar": 0.8346488675360422,
"val_loss": 0.18322986739124367,
"val_weighted_f1": 0.9378908809420781,
"updates": 552
},
{
"epoch": 118,
"train_loss": 0.1301715045130683,
"val_war": 0.938622754491018,
"val_uar": 0.8254178574742898,
"val_loss": 0.1822983456883364,
"val_weighted_f1": 0.9371398367088126,
"updates": 552
},
{
"epoch": 119,
"train_loss": 0.13018532919854012,
"val_war": 0.938622754491018,
"val_uar": 0.8274954948932551,
"val_loss": 0.18285892469559362,
"val_weighted_f1": 0.9372411415559952,
"updates": 552
},
{
"epoch": 120,
"train_loss": 0.12974648134948905,
"val_war": 0.937624750499002,
"val_uar": 0.8244355309556717,
"val_loss": 0.1837939127179201,
"val_weighted_f1": 0.9361600602708166,
"updates": 552
}
],
"test": {
"n": 2047,
"labels_scored": [
0,
1,
2,
3,
4,
5
],
"accuracy": 0.93209574987787,
"uar": 0.8140172329205021,
"macro_f1": 0.8289812043522109,
"weighted_f1": 0.931335012523836,
"per_class": {
"angry": {
"precision": 0.8936170212765957,
"recall": 0.9230769230769231,
"f1": 0.9081081081081082,
"support": 273
},
"disgust": {
"precision": 0.6666666666666666,
"recall": 0.4444444444444444,
"f1": 0.5333333333333333,
"support": 18
},
"fear": {
"precision": 0.7307692307692307,
"recall": 0.6867469879518072,
"f1": 0.7080745341614907,
"support": 83
},
"happy": {
"precision": 0.9740990990990991,
"recall": 0.9686450167973124,
"f1": 0.9713644020213363,
"support": 893
},
"sad": {
"precision": 0.9240837696335078,
"recall": 0.9192708333333334,
"f1": 0.9216710182767625,
"support": 384
},
"surprise": {
"precision": 0.9209876543209876,
"recall": 0.9419191919191919,
"f1": 0.9313358302122348,
"support": 396
}
},
"confusion_matrix": [
[
252,
1,
3,
8,
6,
3
],
[
6,
8,
0,
1,
1,
2
],
[
2,
0,
57,
0,
9,
15
],
[
8,
0,
0,
865,
10,
10
],
[
12,
2,
8,
7,
353,
2
],
[
2,
1,
10,
7,
3,
373
]
],
"report": " precision recall f1-score support\n\n angry 0.89 0.92 0.91 273\n disgust 0.67 0.44 0.53 18\n fear 0.73 0.69 0.71 83\n happy 0.97 0.97 0.97 893\n sad 0.92 0.92 0.92 384\n surprise 0.92 0.94 0.93 396\n\n accuracy 0.93 2047\n macro avg 0.85 0.81 0.83 2047\nweighted avg 0.93 0.93 0.93 2047\n"
},
"test_evaluation_time_sec": 14.172675371170044,
"test_evaluation_peak_mem_gb": 0.562807559967041
}