{ "best_metric": 0.6882572227399814, "best_model_checkpoint": "microsoft/swin-base-patch4-window7-224-finetuned-dsc/checkpoint-336", "epoch": 6.973977695167286, "eval_steps": 500, "global_step": 469, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.14869888475836432, "grad_norm": 3.660515308380127, "learning_rate": 2.1276595744680853e-06, "loss": 0.5513, "step": 10 }, { "epoch": 0.29739776951672864, "grad_norm": 6.621876239776611, "learning_rate": 4.255319148936171e-06, "loss": 0.5441, "step": 20 }, { "epoch": 0.44609665427509293, "grad_norm": 3.9983763694763184, "learning_rate": 6.382978723404256e-06, "loss": 0.514, "step": 30 }, { "epoch": 0.5947955390334573, "grad_norm": 3.805126905441284, "learning_rate": 8.510638297872341e-06, "loss": 0.5214, "step": 40 }, { "epoch": 0.7434944237918215, "grad_norm": 22.76973533630371, "learning_rate": 9.928909952606636e-06, "loss": 0.5192, "step": 50 }, { "epoch": 0.8921933085501859, "grad_norm": 4.288576126098633, "learning_rate": 9.691943127962086e-06, "loss": 0.516, "step": 60 }, { "epoch": 0.9962825278810409, "eval_f1": 0.6873252562907736, "eval_loss": 0.596564769744873, "eval_runtime": 41.8919, "eval_samples_per_second": 51.227, "eval_steps_per_second": 1.623, "step": 67 }, { "epoch": 1.0408921933085502, "grad_norm": 5.175764083862305, "learning_rate": 9.454976303317538e-06, "loss": 0.5326, "step": 70 }, { "epoch": 1.1895910780669146, "grad_norm": 6.2289018630981445, "learning_rate": 9.218009478672988e-06, "loss": 0.5109, "step": 80 }, { "epoch": 1.3382899628252787, "grad_norm": 4.065168857574463, "learning_rate": 8.981042654028437e-06, "loss": 0.4875, "step": 90 }, { "epoch": 1.486988847583643, "grad_norm": 6.063797950744629, "learning_rate": 8.744075829383887e-06, "loss": 0.5308, "step": 100 }, { "epoch": 1.6356877323420074, "grad_norm": 7.436602592468262, "learning_rate": 8.507109004739337e-06, "loss": 0.5143, "step": 110 }, { "epoch": 1.7843866171003717, "grad_norm": 6.663632869720459, "learning_rate": 8.270142180094787e-06, "loss": 0.4992, "step": 120 }, { "epoch": 1.933085501858736, "grad_norm": 5.46168327331543, "learning_rate": 8.033175355450237e-06, "loss": 0.5235, "step": 130 }, { "epoch": 1.9925650557620818, "eval_f1": 0.684995340167754, "eval_loss": 0.5976402163505554, "eval_runtime": 42.0403, "eval_samples_per_second": 51.046, "eval_steps_per_second": 1.617, "step": 134 }, { "epoch": 2.0817843866171004, "grad_norm": 15.794987678527832, "learning_rate": 7.796208530805689e-06, "loss": 0.5037, "step": 140 }, { "epoch": 2.2304832713754648, "grad_norm": 5.5622968673706055, "learning_rate": 7.559241706161138e-06, "loss": 0.5009, "step": 150 }, { "epoch": 2.379182156133829, "grad_norm": 7.298517227172852, "learning_rate": 7.322274881516588e-06, "loss": 0.4899, "step": 160 }, { "epoch": 2.5278810408921935, "grad_norm": 8.912810325622559, "learning_rate": 7.085308056872039e-06, "loss": 0.5268, "step": 170 }, { "epoch": 2.6765799256505574, "grad_norm": 6.775835037231445, "learning_rate": 6.848341232227489e-06, "loss": 0.5132, "step": 180 }, { "epoch": 2.825278810408922, "grad_norm": 5.441810131072998, "learning_rate": 6.611374407582939e-06, "loss": 0.495, "step": 190 }, { "epoch": 2.973977695167286, "grad_norm": 16.180599212646484, "learning_rate": 6.374407582938389e-06, "loss": 0.5327, "step": 200 }, { "epoch": 2.9888475836431225, "eval_f1": 0.6873252562907736, "eval_loss": 0.5961228609085083, "eval_runtime": 41.9926, "eval_samples_per_second": 51.104, "eval_steps_per_second": 1.619, "step": 201 }, { "epoch": 3.1226765799256504, "grad_norm": 6.712996482849121, "learning_rate": 6.137440758293839e-06, "loss": 0.4957, "step": 210 }, { "epoch": 3.2713754646840147, "grad_norm": 5.289604663848877, "learning_rate": 5.90047393364929e-06, "loss": 0.4931, "step": 220 }, { "epoch": 3.420074349442379, "grad_norm": 5.065057754516602, "learning_rate": 5.66350710900474e-06, "loss": 0.5049, "step": 230 }, { "epoch": 3.5687732342007434, "grad_norm": 5.3732757568359375, "learning_rate": 5.42654028436019e-06, "loss": 0.484, "step": 240 }, { "epoch": 3.717472118959108, "grad_norm": 5.8500800132751465, "learning_rate": 5.18957345971564e-06, "loss": 0.5327, "step": 250 }, { "epoch": 3.866171003717472, "grad_norm": 5.2958831787109375, "learning_rate": 4.952606635071091e-06, "loss": 0.5494, "step": 260 }, { "epoch": 4.0, "eval_f1": 0.684995340167754, "eval_loss": 0.5995984673500061, "eval_runtime": 42.2701, "eval_samples_per_second": 50.769, "eval_steps_per_second": 1.609, "step": 269 }, { "epoch": 4.014869888475836, "grad_norm": 5.300852298736572, "learning_rate": 4.715639810426541e-06, "loss": 0.5221, "step": 270 }, { "epoch": 4.163568773234201, "grad_norm": 9.723475456237793, "learning_rate": 4.478672985781991e-06, "loss": 0.5011, "step": 280 }, { "epoch": 4.312267657992565, "grad_norm": 6.0906782150268555, "learning_rate": 4.2417061611374415e-06, "loss": 0.5186, "step": 290 }, { "epoch": 4.4609665427509295, "grad_norm": 9.154439926147461, "learning_rate": 4.004739336492891e-06, "loss": 0.5182, "step": 300 }, { "epoch": 4.609665427509293, "grad_norm": 8.661027908325195, "learning_rate": 3.7677725118483417e-06, "loss": 0.5084, "step": 310 }, { "epoch": 4.758364312267658, "grad_norm": 4.503110885620117, "learning_rate": 3.5308056872037916e-06, "loss": 0.5281, "step": 320 }, { "epoch": 4.907063197026022, "grad_norm": 9.175127983093262, "learning_rate": 3.293838862559242e-06, "loss": 0.5238, "step": 330 }, { "epoch": 4.996282527881041, "eval_f1": 0.6882572227399814, "eval_loss": 0.5969865322113037, "eval_runtime": 42.0429, "eval_samples_per_second": 51.043, "eval_steps_per_second": 1.617, "step": 336 }, { "epoch": 5.055762081784387, "grad_norm": 5.675656795501709, "learning_rate": 3.0568720379146923e-06, "loss": 0.5225, "step": 340 }, { "epoch": 5.204460966542751, "grad_norm": 8.187640190124512, "learning_rate": 2.8199052132701426e-06, "loss": 0.5238, "step": 350 }, { "epoch": 5.353159851301116, "grad_norm": 9.604748725891113, "learning_rate": 2.5829383886255925e-06, "loss": 0.524, "step": 360 }, { "epoch": 5.5018587360594795, "grad_norm": 10.201473236083984, "learning_rate": 2.345971563981043e-06, "loss": 0.5271, "step": 370 }, { "epoch": 5.650557620817844, "grad_norm": 5.156704902648926, "learning_rate": 2.109004739336493e-06, "loss": 0.524, "step": 380 }, { "epoch": 5.799256505576208, "grad_norm": 4.2334818840026855, "learning_rate": 1.8720379146919433e-06, "loss": 0.5219, "step": 390 }, { "epoch": 5.947955390334572, "grad_norm": 10.407471656799316, "learning_rate": 1.6350710900473934e-06, "loss": 0.522, "step": 400 }, { "epoch": 5.992565055762082, "eval_f1": 0.6812674743709226, "eval_loss": 0.6023790240287781, "eval_runtime": 41.9596, "eval_samples_per_second": 51.144, "eval_steps_per_second": 1.621, "step": 403 }, { "epoch": 6.096654275092937, "grad_norm": 5.082653045654297, "learning_rate": 1.3981042654028437e-06, "loss": 0.508, "step": 410 }, { "epoch": 6.245353159851301, "grad_norm": 5.41875696182251, "learning_rate": 1.161137440758294e-06, "loss": 0.5053, "step": 420 }, { "epoch": 6.394052044609666, "grad_norm": 6.110033988952637, "learning_rate": 9.241706161137441e-07, "loss": 0.5157, "step": 430 }, { "epoch": 6.5427509293680295, "grad_norm": 8.269786834716797, "learning_rate": 6.872037914691944e-07, "loss": 0.5157, "step": 440 }, { "epoch": 6.691449814126394, "grad_norm": 6.4608588218688965, "learning_rate": 4.502369668246446e-07, "loss": 0.5176, "step": 450 }, { "epoch": 6.840148698884758, "grad_norm": 7.629857063293457, "learning_rate": 2.132701421800948e-07, "loss": 0.5134, "step": 460 }, { "epoch": 6.973977695167286, "eval_f1": 0.6868592730661697, "eval_loss": 0.5974318981170654, "eval_runtime": 42.3025, "eval_samples_per_second": 50.73, "eval_steps_per_second": 1.607, "step": 469 }, { "epoch": 6.973977695167286, "step": 469, "total_flos": 4.691018488784044e+18, "train_loss": 0.5164371167164622, "train_runtime": 2212.4689, "train_samples_per_second": 27.152, "train_steps_per_second": 0.212 } ], "logging_steps": 10, "max_steps": 469, "num_input_tokens_seen": 0, "num_train_epochs": 7, "save_steps": 500, "stateful_callbacks": { "TrainerControl": { "args": { "should_epoch_stop": false, "should_evaluate": false, "should_log": false, "should_save": true, "should_training_stop": true }, "attributes": {} } }, "total_flos": 4.691018488784044e+18, "train_batch_size": 32, "trial_name": null, "trial_params": null }