| { | |
| "best_metric": 0.9180412395863843, | |
| "best_model_checkpoint": "/home/hoan/projects/nsfw/models/efficientvit_l1.r224_in1k/run-cpz6d1iz/checkpoint-616", | |
| "epoch": 29.647058823529413, | |
| "eval_steps": 500, | |
| "global_step": 630, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 9.337297572513965e-06, | |
| "loss": 1.4974, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.8245565410199557, | |
| "eval_f1": 0.7504652622410692, | |
| "eval_loss": 1.1778662204742432, | |
| "eval_precision": 0.7010977125485033, | |
| "eval_recall": 0.8245565410199557, | |
| "eval_runtime": 7.9368, | |
| "eval_samples_per_second": 454.594, | |
| "eval_steps_per_second": 3.654, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.867459514502793e-05, | |
| "loss": 1.2529, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 2.8011892717541894e-05, | |
| "loss": 0.9843, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_accuracy": 0.8323170731707317, | |
| "eval_f1": 0.7607906844449553, | |
| "eval_loss": 0.9312257170677185, | |
| "eval_precision": 0.7813901035438382, | |
| "eval_recall": 0.8323170731707317, | |
| "eval_runtime": 7.8479, | |
| "eval_samples_per_second": 459.738, | |
| "eval_steps_per_second": 3.695, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.734919029005586e-05, | |
| "loss": 0.9036, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_accuracy": 0.8852549889135255, | |
| "eval_f1": 0.8727382252421404, | |
| "eval_loss": 0.8863803744316101, | |
| "eval_precision": 0.8818471384989105, | |
| "eval_recall": 0.8852549889135255, | |
| "eval_runtime": 8.2029, | |
| "eval_samples_per_second": 439.844, | |
| "eval_steps_per_second": 3.535, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 4.668648786256983e-05, | |
| "loss": 0.8702, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.67, | |
| "learning_rate": 5.602378543508379e-05, | |
| "loss": 0.8272, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9088137472283814, | |
| "eval_f1": 0.9016044442281315, | |
| "eval_loss": 0.827029824256897, | |
| "eval_precision": 0.8977767175054678, | |
| "eval_recall": 0.9088137472283814, | |
| "eval_runtime": 8.6469, | |
| "eval_samples_per_second": 417.259, | |
| "eval_steps_per_second": 3.354, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 6.536108300759776e-05, | |
| "loss": 0.8124, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 4.89, | |
| "learning_rate": 7.469838058011172e-05, | |
| "loss": 0.791, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.8896895787139689, | |
| "eval_f1": 0.8908302243934083, | |
| "eval_loss": 0.8426142930984497, | |
| "eval_precision": 0.8952544356380175, | |
| "eval_recall": 0.8896895787139689, | |
| "eval_runtime": 7.9673, | |
| "eval_samples_per_second": 452.854, | |
| "eval_steps_per_second": 3.64, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 5.51, | |
| "learning_rate": 8.403567815262568e-05, | |
| "loss": 0.7742, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_accuracy": 0.9077050997782705, | |
| "eval_f1": 0.8952158275122053, | |
| "eval_loss": 0.8156972527503967, | |
| "eval_precision": 0.900683213811247, | |
| "eval_recall": 0.9077050997782705, | |
| "eval_runtime": 8.0605, | |
| "eval_samples_per_second": 447.614, | |
| "eval_steps_per_second": 3.598, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 6.12, | |
| "learning_rate": 9.048589657902734e-05, | |
| "loss": 0.7771, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.73, | |
| "learning_rate": 9.02461453820045e-05, | |
| "loss": 0.7696, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.9149113082039911, | |
| "eval_f1": 0.9089808097871447, | |
| "eval_loss": 0.8019198775291443, | |
| "eval_precision": 0.9083537029648736, | |
| "eval_recall": 0.9149113082039911, | |
| "eval_runtime": 8.2664, | |
| "eval_samples_per_second": 436.464, | |
| "eval_steps_per_second": 3.508, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 7.34, | |
| "learning_rate": 8.971109434044089e-05, | |
| "loss": 0.7627, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 7.95, | |
| "learning_rate": 8.888425487558681e-05, | |
| "loss": 0.7579, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9099223946784922, | |
| "eval_f1": 0.9056360787135247, | |
| "eval_loss": 0.8119347095489502, | |
| "eval_precision": 0.9037596807714011, | |
| "eval_recall": 0.9099223946784922, | |
| "eval_runtime": 8.3286, | |
| "eval_samples_per_second": 433.208, | |
| "eval_steps_per_second": 3.482, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 8.777105335136256e-05, | |
| "loss": 0.753, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.9143569844789357, | |
| "eval_f1": 0.905016737540752, | |
| "eval_loss": 0.8058480024337769, | |
| "eval_precision": 0.9043267295510783, | |
| "eval_recall": 0.9143569844789357, | |
| "eval_runtime": 7.8299, | |
| "eval_samples_per_second": 460.796, | |
| "eval_steps_per_second": 3.704, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 9.18, | |
| "learning_rate": 8.637879546233632e-05, | |
| "loss": 0.7467, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 9.79, | |
| "learning_rate": 8.471661828807179e-05, | |
| "loss": 0.7414, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 9.98, | |
| "eval_accuracy": 0.8910753880266076, | |
| "eval_f1": 0.8925899353200606, | |
| "eval_loss": 0.8334646821022034, | |
| "eval_precision": 0.8965909503064864, | |
| "eval_recall": 0.8910753880266076, | |
| "eval_runtime": 8.1268, | |
| "eval_samples_per_second": 443.961, | |
| "eval_steps_per_second": 3.568, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 8.279543032850277e-05, | |
| "loss": 0.736, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_accuracy": 0.899390243902439, | |
| "eval_f1": 0.9002456226563228, | |
| "eval_loss": 0.8323202133178711, | |
| "eval_precision": 0.903732584843102, | |
| "eval_recall": 0.899390243902439, | |
| "eval_runtime": 7.848, | |
| "eval_samples_per_second": 459.738, | |
| "eval_steps_per_second": 3.695, | |
| "step": 233 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 8.062783991386883e-05, | |
| "loss": 0.7351, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 11.62, | |
| "learning_rate": 7.822807245904186e-05, | |
| "loss": 0.7312, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9154656319290465, | |
| "eval_f1": 0.9086965051403931, | |
| "eval_loss": 0.802578330039978, | |
| "eval_precision": 0.9043713569516466, | |
| "eval_recall": 0.9154656319290465, | |
| "eval_runtime": 7.9576, | |
| "eval_samples_per_second": 453.405, | |
| "eval_steps_per_second": 3.644, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 12.24, | |
| "learning_rate": 7.561187710528515e-05, | |
| "loss": 0.7311, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.85, | |
| "learning_rate": 7.279642336213425e-05, | |
| "loss": 0.722, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.9054878048780488, | |
| "eval_f1": 0.9053749893268752, | |
| "eval_loss": 0.8218502402305603, | |
| "eval_precision": 0.9072092893262743, | |
| "eval_recall": 0.9054878048780488, | |
| "eval_runtime": 8.5776, | |
| "eval_samples_per_second": 420.629, | |
| "eval_steps_per_second": 3.381, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 13.46, | |
| "learning_rate": 6.980018842771594e-05, | |
| "loss": 0.728, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 13.98, | |
| "eval_accuracy": 0.9146341463414634, | |
| "eval_f1": 0.9088337129199915, | |
| "eval_loss": 0.803844690322876, | |
| "eval_precision": 0.9053003849300292, | |
| "eval_recall": 0.9146341463414634, | |
| "eval_runtime": 8.4963, | |
| "eval_samples_per_second": 424.653, | |
| "eval_steps_per_second": 3.413, | |
| "step": 297 | |
| }, | |
| { | |
| "epoch": 14.07, | |
| "learning_rate": 6.664283592699743e-05, | |
| "loss": 0.7174, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 14.68, | |
| "learning_rate": 6.33450868637792e-05, | |
| "loss": 0.7178, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 14.96, | |
| "eval_accuracy": 0.917960088691796, | |
| "eval_f1": 0.9115339219530667, | |
| "eval_loss": 0.8020056486129761, | |
| "eval_precision": 0.9126518511003248, | |
| "eval_recall": 0.917960088691796, | |
| "eval_runtime": 8.0058, | |
| "eval_samples_per_second": 450.671, | |
| "eval_steps_per_second": 3.622, | |
| "step": 318 | |
| }, | |
| { | |
| "epoch": 15.29, | |
| "learning_rate": 5.99285836333455e-05, | |
| "loss": 0.7168, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 15.91, | |
| "learning_rate": 5.6415747988226656e-05, | |
| "loss": 0.7181, | |
| "step": 338 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9168514412416852, | |
| "eval_f1": 0.9074858536331818, | |
| "eval_loss": 0.8075320720672607, | |
| "eval_precision": 0.9090067831913919, | |
| "eval_recall": 0.9168514412416852, | |
| "eval_runtime": 8.2679, | |
| "eval_samples_per_second": 436.387, | |
| "eval_steps_per_second": 3.508, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.52, | |
| "learning_rate": 5.282963388921312e-05, | |
| "loss": 0.7132, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.9115853658536586, | |
| "eval_f1": 0.9105576927569372, | |
| "eval_loss": 0.8095524311065674, | |
| "eval_precision": 0.9105897373648217, | |
| "eval_recall": 0.9115853658536586, | |
| "eval_runtime": 8.3192, | |
| "eval_samples_per_second": 433.694, | |
| "eval_steps_per_second": 3.486, | |
| "step": 361 | |
| }, | |
| { | |
| "epoch": 17.13, | |
| "learning_rate": 4.919377620732666e-05, | |
| "loss": 0.7122, | |
| "step": 364 | |
| }, | |
| { | |
| "epoch": 17.74, | |
| "learning_rate": 4.553203626968464e-05, | |
| "loss": 0.7102, | |
| "step": 377 | |
| }, | |
| { | |
| "epoch": 17.98, | |
| "eval_accuracy": 0.9126940133037694, | |
| "eval_f1": 0.9109325265921849, | |
| "eval_loss": 0.8063324093818665, | |
| "eval_precision": 0.909426612792062, | |
| "eval_recall": 0.9126940133037694, | |
| "eval_runtime": 8.1988, | |
| "eval_samples_per_second": 440.063, | |
| "eval_steps_per_second": 3.537, | |
| "step": 382 | |
| }, | |
| { | |
| "epoch": 18.35, | |
| "learning_rate": 4.1868445262905336e-05, | |
| "loss": 0.7107, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "learning_rate": 3.8227046521762955e-05, | |
| "loss": 0.7057, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 18.96, | |
| "eval_accuracy": 0.9124168514412417, | |
| "eval_f1": 0.9116152740130952, | |
| "eval_loss": 0.8127285838127136, | |
| "eval_precision": 0.9140500451639162, | |
| "eval_recall": 0.9124168514412417, | |
| "eval_runtime": 8.2949, | |
| "eval_samples_per_second": 434.968, | |
| "eval_steps_per_second": 3.496, | |
| "step": 403 | |
| }, | |
| { | |
| "epoch": 19.58, | |
| "learning_rate": 3.463173773811715e-05, | |
| "loss": 0.7102, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9193458980044346, | |
| "eval_f1": 0.9124581241091632, | |
| "eval_loss": 0.8046479225158691, | |
| "eval_precision": 0.9094511683912293, | |
| "eval_recall": 0.9193458980044346, | |
| "eval_runtime": 8.4514, | |
| "eval_samples_per_second": 426.911, | |
| "eval_steps_per_second": 3.431, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 20.19, | |
| "learning_rate": 3.1106114125664475e-05, | |
| "loss": 0.7031, | |
| "step": 429 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 2.7673313569787155e-05, | |
| "loss": 0.6998, | |
| "step": 442 | |
| }, | |
| { | |
| "epoch": 20.99, | |
| "eval_accuracy": 0.916019955654102, | |
| "eval_f1": 0.9107592161617032, | |
| "eval_loss": 0.8098756074905396, | |
| "eval_precision": 0.9094635680966053, | |
| "eval_recall": 0.916019955654102, | |
| "eval_runtime": 8.2362, | |
| "eval_samples_per_second": 438.069, | |
| "eval_steps_per_second": 3.521, | |
| "step": 446 | |
| }, | |
| { | |
| "epoch": 21.41, | |
| "learning_rate": 2.4355864778745852e-05, | |
| "loss": 0.7051, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 21.98, | |
| "eval_accuracy": 0.9174057649667405, | |
| "eval_f1": 0.9142268993626627, | |
| "eval_loss": 0.8050925731658936, | |
| "eval_precision": 0.9130379000974386, | |
| "eval_recall": 0.9174057649667405, | |
| "eval_runtime": 8.3981, | |
| "eval_samples_per_second": 429.62, | |
| "eval_steps_per_second": 3.453, | |
| "step": 467 | |
| }, | |
| { | |
| "epoch": 22.02, | |
| "learning_rate": 2.1175539432766786e-05, | |
| "loss": 0.7013, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 22.64, | |
| "learning_rate": 1.8153209301335927e-05, | |
| "loss": 0.6986, | |
| "step": 481 | |
| }, | |
| { | |
| "epoch": 22.96, | |
| "eval_accuracy": 0.9143569844789357, | |
| "eval_f1": 0.9094601509445771, | |
| "eval_loss": 0.8089693188667297, | |
| "eval_precision": 0.9069934469200637, | |
| "eval_recall": 0.9143569844789357, | |
| "eval_runtime": 8.7255, | |
| "eval_samples_per_second": 413.501, | |
| "eval_steps_per_second": 3.324, | |
| "step": 488 | |
| }, | |
| { | |
| "epoch": 23.25, | |
| "learning_rate": 1.5308709266407565e-05, | |
| "loss": 0.6993, | |
| "step": 494 | |
| }, | |
| { | |
| "epoch": 23.86, | |
| "learning_rate": 1.2660707150475642e-05, | |
| "loss": 0.7004, | |
| "step": 507 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9190687361419069, | |
| "eval_f1": 0.9157749931932022, | |
| "eval_loss": 0.7969859838485718, | |
| "eval_precision": 0.9146257718265792, | |
| "eval_recall": 0.9190687361419069, | |
| "eval_runtime": 7.8691, | |
| "eval_samples_per_second": 458.501, | |
| "eval_steps_per_second": 3.685, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 24.47, | |
| "learning_rate": 1.0226581203797468e-05, | |
| "loss": 0.6992, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 24.99, | |
| "eval_accuracy": 0.9187915742793792, | |
| "eval_f1": 0.9128722156884925, | |
| "eval_loss": 0.8054778575897217, | |
| "eval_precision": 0.9123823991036795, | |
| "eval_recall": 0.9187915742793792, | |
| "eval_runtime": 8.3104, | |
| "eval_samples_per_second": 434.157, | |
| "eval_steps_per_second": 3.49, | |
| "step": 531 | |
| }, | |
| { | |
| "epoch": 25.08, | |
| "learning_rate": 8.022306054793715e-06, | |
| "loss": 0.6988, | |
| "step": 533 | |
| }, | |
| { | |
| "epoch": 25.69, | |
| "learning_rate": 6.062347872107096e-06, | |
| "loss": 0.6956, | |
| "step": 546 | |
| }, | |
| { | |
| "epoch": 25.98, | |
| "eval_accuracy": 0.9201773835920177, | |
| "eval_f1": 0.9159544845078982, | |
| "eval_loss": 0.7970059514045715, | |
| "eval_precision": 0.9139707421362406, | |
| "eval_recall": 0.9201773835920177, | |
| "eval_runtime": 8.3576, | |
| "eval_samples_per_second": 431.703, | |
| "eval_steps_per_second": 3.47, | |
| "step": 552 | |
| }, | |
| { | |
| "epoch": 26.31, | |
| "learning_rate": 4.359569426347826e-06, | |
| "loss": 0.6955, | |
| "step": 559 | |
| }, | |
| { | |
| "epoch": 26.92, | |
| "learning_rate": 2.925145674584495e-06, | |
| "loss": 0.6941, | |
| "step": 572 | |
| }, | |
| { | |
| "epoch": 26.96, | |
| "eval_accuracy": 0.9193458980044346, | |
| "eval_f1": 0.9158537735891641, | |
| "eval_loss": 0.8026701807975769, | |
| "eval_precision": 0.9141634537835975, | |
| "eval_recall": 0.9193458980044346, | |
| "eval_runtime": 8.2835, | |
| "eval_samples_per_second": 435.562, | |
| "eval_steps_per_second": 3.501, | |
| "step": 573 | |
| }, | |
| { | |
| "epoch": 27.53, | |
| "learning_rate": 1.7684904215807802e-06, | |
| "loss": 0.6973, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9149113082039911, | |
| "eval_f1": 0.9105265775058081, | |
| "eval_loss": 0.8031827807426453, | |
| "eval_precision": 0.9092055837026697, | |
| "eval_recall": 0.9149113082039911, | |
| "eval_runtime": 8.0208, | |
| "eval_samples_per_second": 449.829, | |
| "eval_steps_per_second": 3.616, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 28.14, | |
| "learning_rate": 8.971945390837938e-07, | |
| "loss": 0.698, | |
| "step": 598 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "learning_rate": 3.169761486173364e-07, | |
| "loss": 0.6969, | |
| "step": 611 | |
| }, | |
| { | |
| "epoch": 28.99, | |
| "eval_accuracy": 0.9223946784922394, | |
| "eval_f1": 0.9180412395863843, | |
| "eval_loss": 0.7970722913742065, | |
| "eval_precision": 0.9152796171817729, | |
| "eval_recall": 0.9223946784922394, | |
| "eval_runtime": 8.4656, | |
| "eval_samples_per_second": 426.196, | |
| "eval_steps_per_second": 3.426, | |
| "step": 616 | |
| }, | |
| { | |
| "epoch": 29.36, | |
| "learning_rate": 3.1643094719280604e-08, | |
| "loss": 0.6981, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 29.65, | |
| "eval_accuracy": 0.9187915742793792, | |
| "eval_f1": 0.9142339823689526, | |
| "eval_loss": 0.804097056388855, | |
| "eval_precision": 0.913375902362099, | |
| "eval_recall": 0.9187915742793792, | |
| "eval_runtime": 7.8761, | |
| "eval_samples_per_second": 458.092, | |
| "eval_steps_per_second": 3.682, | |
| "step": 630 | |
| } | |
| ], | |
| "logging_steps": 13, | |
| "max_steps": 630, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 0.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "label_smoothing_factor": 0.029883991666275297, | |
| "learning_rate": 9.049996108744304e-05, | |
| "metric": "eval/loss", | |
| "weight_decay": 0.011787322184488629 | |
| } | |
| } | |