| { | |
| "best_metric": 0.8349146110056926, | |
| "best_model_checkpoint": "swin-base-patch4-window7-224-in22k-MM/NMM_Classification_base_V10/checkpoint-81", | |
| "epoch": 6.892307692307693, | |
| "eval_steps": 500, | |
| "global_step": 112, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 3.4730300903320312, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.8416, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.9846153846153847, | |
| "step": 16, | |
| "train_accuracy": 0.7838165716365401, | |
| "train_auc_class_0": 0.9185630172913702, | |
| "train_auc_class_1": 0.8867337604929844, | |
| "train_auc_class_2": 0.8750511256900684, | |
| "train_auc_overall": 0.8934493011581409, | |
| "train_loss": 0.5212814211845398, | |
| "train_runtime": 738.2478, | |
| "train_samples_per_second": 11.166, | |
| "train_steps_per_second": 0.088 | |
| }, | |
| { | |
| "epoch": 0.9846153846153847, | |
| "eval_accuracy": 0.7751423149905123, | |
| "eval_auc_class_0": 0.9351696867321869, | |
| "eval_auc_class_1": 0.9013257754334977, | |
| "eval_auc_class_2": 0.896788439192459, | |
| "eval_auc_overall": 0.9110946337860479, | |
| "eval_loss": 0.5271921157836914, | |
| "eval_runtime": 226.0415, | |
| "eval_samples_per_second": 9.326, | |
| "eval_steps_per_second": 0.075, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 2.644275426864624, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 0.5698, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 2.8130509853363037, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.4812, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9692307692307693, | |
| "step": 32, | |
| "train_accuracy": 0.8125682397185491, | |
| "train_auc_class_0": 0.9453865517423163, | |
| "train_auc_class_1": 0.9307891812846302, | |
| "train_auc_class_2": 0.9068255493827921, | |
| "train_auc_overall": 0.9276670941365796, | |
| "train_loss": 0.45027220249176025, | |
| "train_runtime": 742.3599, | |
| "train_samples_per_second": 11.104, | |
| "train_steps_per_second": 0.088 | |
| }, | |
| { | |
| "epoch": 1.9692307692307693, | |
| "eval_accuracy": 0.8088235294117647, | |
| "eval_auc_class_0": 0.9525644963144962, | |
| "eval_auc_class_1": 0.92725521408195, | |
| "eval_auc_class_2": 0.9013002141426902, | |
| "eval_auc_overall": 0.9270399748463788, | |
| "eval_loss": 0.4628311097621918, | |
| "eval_runtime": 227.1982, | |
| "eval_samples_per_second": 9.278, | |
| "eval_steps_per_second": 0.075, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 4.537336349487305, | |
| "learning_rate": 3.6e-05, | |
| "loss": 0.4444, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.953846153846154, | |
| "step": 48, | |
| "train_accuracy": 0.8330704840470703, | |
| "train_auc_class_0": 0.9573463611682219, | |
| "train_auc_class_1": 0.9445183457629789, | |
| "train_auc_class_2": 0.9234336270362926, | |
| "train_auc_overall": 0.9417661113224979, | |
| "train_loss": 0.3945094048976898, | |
| "train_runtime": 749.6504, | |
| "train_samples_per_second": 10.996, | |
| "train_steps_per_second": 0.087 | |
| }, | |
| { | |
| "epoch": 2.953846153846154, | |
| "eval_accuracy": 0.8273244781783681, | |
| "eval_auc_class_0": 0.9561783374283375, | |
| "eval_auc_class_1": 0.9278154008848364, | |
| "eval_auc_class_2": 0.9137749676178002, | |
| "eval_auc_overall": 0.932589568643658, | |
| "eval_loss": 0.42404672503471375, | |
| "eval_runtime": 226.6025, | |
| "eval_samples_per_second": 9.303, | |
| "eval_steps_per_second": 0.075, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.965461254119873, | |
| "learning_rate": 3.1e-05, | |
| "loss": 0.4265, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 3.6663880348205566, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 0.4055, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 65, | |
| "train_accuracy": 0.8460511949532937, | |
| "train_auc_class_0": 0.9634787762270021, | |
| "train_auc_class_1": 0.9469802033822643, | |
| "train_auc_class_2": 0.9308556192528558, | |
| "train_auc_overall": 0.9471048662873741, | |
| "train_loss": 0.37029632925987244, | |
| "train_runtime": 746.5442, | |
| "train_samples_per_second": 11.042, | |
| "train_steps_per_second": 0.087 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8287476280834914, | |
| "eval_auc_class_0": 0.9613994676494677, | |
| "eval_auc_class_1": 0.9314889691268666, | |
| "eval_auc_class_2": 0.9206087950151214, | |
| "eval_auc_overall": 0.9378324105971519, | |
| "eval_loss": 0.41696038842201233, | |
| "eval_runtime": 226.2202, | |
| "eval_samples_per_second": 9.318, | |
| "eval_steps_per_second": 0.075, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 3.821587324142456, | |
| "learning_rate": 2.1e-05, | |
| "loss": 0.3875, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 3.1884751319885254, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.3914, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.984615384615385, | |
| "step": 81, | |
| "train_accuracy": 0.8536940434307898, | |
| "train_auc_class_0": 0.9671529040803416, | |
| "train_auc_class_1": 0.9496355903643986, | |
| "train_auc_class_2": 0.9367910485339515, | |
| "train_auc_overall": 0.9511931809928972, | |
| "train_loss": 0.3558380603790283, | |
| "train_runtime": 752.3511, | |
| "train_samples_per_second": 10.956, | |
| "train_steps_per_second": 0.086 | |
| }, | |
| { | |
| "epoch": 4.984615384615385, | |
| "eval_accuracy": 0.8349146110056926, | |
| "eval_auc_class_0": 0.9613559582309583, | |
| "eval_auc_class_1": 0.9315019107361742, | |
| "eval_auc_class_2": 0.920726636279075, | |
| "eval_auc_overall": 0.9378615017487358, | |
| "eval_loss": 0.40655088424682617, | |
| "eval_runtime": 227.6818, | |
| "eval_samples_per_second": 9.259, | |
| "eval_steps_per_second": 0.075, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 5.538461538461538, | |
| "grad_norm": 2.0869638919830322, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 0.3731, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.969230769230769, | |
| "step": 97, | |
| "train_accuracy": 0.8507824821060294, | |
| "train_auc_class_0": 0.9680180491855133, | |
| "train_auc_class_1": 0.9494889730661834, | |
| "train_auc_class_2": 0.9379508502279332, | |
| "train_auc_overall": 0.9518192908265433, | |
| "train_loss": 0.35146117210388184, | |
| "train_runtime": 745.296, | |
| "train_samples_per_second": 11.06, | |
| "train_steps_per_second": 0.087 | |
| }, | |
| { | |
| "epoch": 5.969230769230769, | |
| "eval_accuracy": 0.8334914611005693, | |
| "eval_auc_class_0": 0.9624193796068795, | |
| "eval_auc_class_1": 0.9319585646645997, | |
| "eval_auc_class_2": 0.9218326055186417, | |
| "eval_auc_overall": 0.9387368499300403, | |
| "eval_loss": 0.40432390570640564, | |
| "eval_runtime": 226.3347, | |
| "eval_samples_per_second": 9.314, | |
| "eval_steps_per_second": 0.075, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 2.3311638832092285, | |
| "learning_rate": 6e-06, | |
| "loss": 0.357, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "grad_norm": 2.3013312816619873, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.3687, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.892307692307693, | |
| "step": 112, | |
| "train_accuracy": 0.8533300982651947, | |
| "train_auc_class_0": 0.9690346749960462, | |
| "train_auc_class_1": 0.9513212789842702, | |
| "train_auc_class_2": 0.9389334461695297, | |
| "train_auc_overall": 0.9530964667166154, | |
| "train_loss": 0.34802600741386414, | |
| "train_runtime": 740.1343, | |
| "train_samples_per_second": 11.137, | |
| "train_steps_per_second": 0.088 | |
| }, | |
| { | |
| "epoch": 6.892307692307693, | |
| "eval_accuracy": 0.8344402277039848, | |
| "eval_auc_class_0": 0.9626446048321048, | |
| "eval_auc_class_1": 0.9327683396469899, | |
| "eval_auc_class_2": 0.9223551639104368, | |
| "eval_auc_overall": 0.9392560361298439, | |
| "eval_loss": 0.40155309438705444, | |
| "eval_runtime": 226.5604, | |
| "eval_samples_per_second": 9.304, | |
| "eval_steps_per_second": 0.075, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 6.892307692307693, | |
| "step": 112, | |
| "total_flos": 4.4565045535859466e+18, | |
| "train_loss": 0.45718009237732204, | |
| "train_runtime": 12452.0227, | |
| "train_samples_per_second": 4.634, | |
| "train_steps_per_second": 0.009 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 112, | |
| "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": 4.4565045535859466e+18, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |