| { | |
| "best_metric": 0.11592021584510803, | |
| "best_model_checkpoint": "/home1/datawork/mcontini/models/multilabel/huggingface/dino-large-2023_12_19-kornia_img-size518_batch-size16_epochs20/checkpoint-9112", | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 10720, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.01, | |
| "loss": 0.2545, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.41336191496963204, | |
| "eval_f1_macro": 0.5552795713944485, | |
| "eval_f1_micro": 0.7016183870529035, | |
| "eval_loss": 0.19357216358184814, | |
| "eval_roc_auc": 0.7908326703460342, | |
| "eval_runtime": 669.1996, | |
| "eval_samples_per_second": 4.183, | |
| "eval_steps_per_second": 0.262, | |
| "learning_rate": 0.01, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.01, | |
| "loss": 0.2163, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4240800285816363, | |
| "eval_f1_macro": 0.6939559668202788, | |
| "eval_f1_micro": 0.7671674437345306, | |
| "eval_loss": 0.16442517936229706, | |
| "eval_roc_auc": 0.8668895881413139, | |
| "eval_runtime": 667.4861, | |
| "eval_samples_per_second": 4.193, | |
| "eval_steps_per_second": 0.262, | |
| "learning_rate": 0.01, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.01, | |
| "loss": 0.2142, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4258663808503037, | |
| "eval_f1_macro": 0.6226385110394398, | |
| "eval_f1_micro": 0.7264495032071437, | |
| "eval_loss": 0.17197823524475098, | |
| "eval_roc_auc": 0.8209912192682602, | |
| "eval_runtime": 671.4241, | |
| "eval_samples_per_second": 4.169, | |
| "eval_steps_per_second": 0.261, | |
| "learning_rate": 0.01, | |
| "step": 1608 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.01, | |
| "loss": 0.2107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.40192926045016075, | |
| "eval_f1_macro": 0.60556232845899, | |
| "eval_f1_micro": 0.7311342592592591, | |
| "eval_loss": 0.17794817686080933, | |
| "eval_roc_auc": 0.844154343443223, | |
| "eval_runtime": 669.3476, | |
| "eval_samples_per_second": 4.182, | |
| "eval_steps_per_second": 0.261, | |
| "learning_rate": 0.01, | |
| "step": 2144 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 0.01, | |
| "loss": 0.2117, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.38335119685602, | |
| "eval_f1_macro": 0.674484878286151, | |
| "eval_f1_micro": 0.754215411654858, | |
| "eval_loss": 0.1834840625524521, | |
| "eval_roc_auc": 0.8723557298678095, | |
| "eval_runtime": 667.4497, | |
| "eval_samples_per_second": 4.194, | |
| "eval_steps_per_second": 0.262, | |
| "learning_rate": 0.01, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.01, | |
| "loss": 0.2171, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.42086459449803504, | |
| "eval_f1_macro": 0.595946612091723, | |
| "eval_f1_micro": 0.7347275031685679, | |
| "eval_loss": 0.1732301414012909, | |
| "eval_roc_auc": 0.8236385276926494, | |
| "eval_runtime": 665.3111, | |
| "eval_samples_per_second": 4.207, | |
| "eval_steps_per_second": 0.263, | |
| "learning_rate": 0.01, | |
| "step": 3216 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.01, | |
| "loss": 0.2178, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.3904966059306895, | |
| "eval_f1_macro": 0.5932113008464752, | |
| "eval_f1_micro": 0.7253084446717382, | |
| "eval_loss": 0.2698315978050232, | |
| "eval_roc_auc": 0.8165478137957464, | |
| "eval_runtime": 666.3656, | |
| "eval_samples_per_second": 4.2, | |
| "eval_steps_per_second": 0.263, | |
| "learning_rate": 0.01, | |
| "step": 3752 | |
| }, | |
| { | |
| "epoch": 7.46, | |
| "learning_rate": 0.01, | |
| "loss": 0.2177, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4119328331546981, | |
| "eval_f1_macro": 0.6280317383192718, | |
| "eval_f1_micro": 0.7360358677377172, | |
| "eval_loss": 0.19401037693023682, | |
| "eval_roc_auc": 0.828564591229026, | |
| "eval_runtime": 671.4879, | |
| "eval_samples_per_second": 4.168, | |
| "eval_steps_per_second": 0.261, | |
| "learning_rate": 0.01, | |
| "step": 4288 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.001, | |
| "loss": 0.212, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.48981779206859594, | |
| "eval_f1_macro": 0.7491476401858057, | |
| "eval_f1_micro": 0.7992611571232794, | |
| "eval_loss": 0.1454659402370453, | |
| "eval_roc_auc": 0.8756991587148717, | |
| "eval_runtime": 673.1035, | |
| "eval_samples_per_second": 4.158, | |
| "eval_steps_per_second": 0.26, | |
| "learning_rate": 0.001, | |
| "step": 4824 | |
| }, | |
| { | |
| "epoch": 9.33, | |
| "learning_rate": 0.001, | |
| "loss": 0.1761, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5123258306538049, | |
| "eval_f1_macro": 0.7661404438716006, | |
| "eval_f1_micro": 0.8116191765647116, | |
| "eval_loss": 0.13573050498962402, | |
| "eval_roc_auc": 0.8733133406359384, | |
| "eval_runtime": 691.4938, | |
| "eval_samples_per_second": 4.048, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 10.26, | |
| "learning_rate": 0.001, | |
| "loss": 0.1681, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.5166130760986066, | |
| "eval_f1_macro": 0.7752662433951194, | |
| "eval_f1_micro": 0.8151612707282998, | |
| "eval_loss": 0.1386074423789978, | |
| "eval_roc_auc": 0.8790914856208012, | |
| "eval_runtime": 689.8939, | |
| "eval_samples_per_second": 4.057, | |
| "eval_steps_per_second": 0.254, | |
| "learning_rate": 0.001, | |
| "step": 5896 | |
| }, | |
| { | |
| "epoch": 11.19, | |
| "learning_rate": 0.001, | |
| "loss": 0.1579, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.5162558056448732, | |
| "eval_f1_macro": 0.7826923679681995, | |
| "eval_f1_micro": 0.8219824282092104, | |
| "eval_loss": 0.1820007711648941, | |
| "eval_roc_auc": 0.8918796772066396, | |
| "eval_runtime": 692.3943, | |
| "eval_samples_per_second": 4.042, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 6432 | |
| }, | |
| { | |
| "epoch": 12.13, | |
| "learning_rate": 0.001, | |
| "loss": 0.1553, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.5326902465166131, | |
| "eval_f1_macro": 0.7907678649401602, | |
| "eval_f1_micro": 0.8296795952782462, | |
| "eval_loss": 0.1227828711271286, | |
| "eval_roc_auc": 0.8898245747393245, | |
| "eval_runtime": 691.2613, | |
| "eval_samples_per_second": 4.049, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 6968 | |
| }, | |
| { | |
| "epoch": 13.06, | |
| "learning_rate": 0.001, | |
| "loss": 0.1517, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 0.001, | |
| "loss": 0.1512, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.5301893533404788, | |
| "eval_f1_macro": 0.7815063263088662, | |
| "eval_f1_micro": 0.8257773991358852, | |
| "eval_loss": 0.12334247678518295, | |
| "eval_roc_auc": 0.8845045355403204, | |
| "eval_runtime": 684.2138, | |
| "eval_samples_per_second": 4.091, | |
| "eval_steps_per_second": 0.256, | |
| "learning_rate": 0.001, | |
| "step": 7504 | |
| }, | |
| { | |
| "epoch": 14.93, | |
| "learning_rate": 0.001, | |
| "loss": 0.1508, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.5305466237942122, | |
| "eval_f1_macro": 0.7681866841890026, | |
| "eval_f1_micro": 0.817921948794432, | |
| "eval_loss": 0.12476785480976105, | |
| "eval_roc_auc": 0.8739908739957604, | |
| "eval_runtime": 690.4773, | |
| "eval_samples_per_second": 4.054, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 8040 | |
| }, | |
| { | |
| "epoch": 15.86, | |
| "learning_rate": 0.001, | |
| "loss": 0.1499, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5394783851375491, | |
| "eval_f1_macro": 0.7902552675483185, | |
| "eval_f1_micro": 0.8276715549780336, | |
| "eval_loss": 0.11932893097400665, | |
| "eval_roc_auc": 0.8805997685784225, | |
| "eval_runtime": 690.469, | |
| "eval_samples_per_second": 4.054, | |
| "eval_steps_per_second": 0.253, | |
| "learning_rate": 0.001, | |
| "step": 8576 | |
| }, | |
| { | |
| "epoch": 16.79, | |
| "learning_rate": 0.001, | |
| "loss": 0.1435, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.5380493033226152, | |
| "eval_f1_macro": 0.7995743847858905, | |
| "eval_f1_micro": 0.8381208681916574, | |
| "eval_loss": 0.11592021584510803, | |
| "eval_roc_auc": 0.9037053380559585, | |
| "eval_runtime": 685.0697, | |
| "eval_samples_per_second": 4.086, | |
| "eval_steps_per_second": 0.255, | |
| "learning_rate": 0.001, | |
| "step": 9112 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 0.001, | |
| "loss": 0.1463, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.5480528760271526, | |
| "eval_f1_macro": 0.803284244956481, | |
| "eval_f1_micro": 0.8392878623842175, | |
| "eval_loss": 0.11661119759082794, | |
| "eval_roc_auc": 0.8956742640228449, | |
| "eval_runtime": 684.4351, | |
| "eval_samples_per_second": 4.09, | |
| "eval_steps_per_second": 0.256, | |
| "learning_rate": 0.001, | |
| "step": 9648 | |
| }, | |
| { | |
| "epoch": 18.66, | |
| "learning_rate": 0.001, | |
| "loss": 0.1423, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.5459092533047517, | |
| "eval_f1_macro": 0.8008505005735617, | |
| "eval_f1_micro": 0.8327005511328843, | |
| "eval_loss": 0.1216365247964859, | |
| "eval_roc_auc": 0.8864852456725716, | |
| "eval_runtime": 683.4387, | |
| "eval_samples_per_second": 4.095, | |
| "eval_steps_per_second": 0.256, | |
| "learning_rate": 0.001, | |
| "step": 10184 | |
| }, | |
| { | |
| "epoch": 19.59, | |
| "learning_rate": 0.001, | |
| "loss": 0.1444, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5509110396570204, | |
| "eval_f1_macro": 0.801973744395412, | |
| "eval_f1_micro": 0.8382846336851095, | |
| "eval_loss": 0.1171422228217125, | |
| "eval_roc_auc": 0.8907672852306303, | |
| "eval_runtime": 675.9942, | |
| "eval_samples_per_second": 4.141, | |
| "eval_steps_per_second": 0.259, | |
| "learning_rate": 0.001, | |
| "step": 10720 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.001, | |
| "step": 10720, | |
| "total_flos": 2.5409000327874334e+20, | |
| "train_loss": 0.18060117693089728, | |
| "train_runtime": 57213.4657, | |
| "train_samples_per_second": 2.995, | |
| "train_steps_per_second": 0.187 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10720, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 2.5409000327874334e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |