| { | |
| "best_metric": 0.8905191873589164, | |
| "best_model_checkpoint": "vit-base-patch16-224-finetuned-brain-tumor-classification/checkpoint-480", | |
| "epoch": 9.896907216494846, | |
| "eval_steps": 500, | |
| "global_step": 480, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.41237113402061853, | |
| "grad_norm": 3.4255573749542236, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 3.6966, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8247422680412371, | |
| "grad_norm": 3.2779064178466797, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 3.1659, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9896907216494846, | |
| "eval_accuracy": 0.40857787810383744, | |
| "eval_loss": 2.4060096740722656, | |
| "eval_runtime": 7.918, | |
| "eval_samples_per_second": 111.897, | |
| "eval_steps_per_second": 3.536, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.2371134020618557, | |
| "grad_norm": 3.0740227699279785, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 2.4212, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6494845360824741, | |
| "grad_norm": 3.2845816612243652, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.8381, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6772009029345373, | |
| "eval_loss": 1.2904105186462402, | |
| "eval_runtime": 8.0965, | |
| "eval_samples_per_second": 109.429, | |
| "eval_steps_per_second": 3.458, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.0618556701030926, | |
| "grad_norm": 2.981458902359009, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 1.4886, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.4742268041237114, | |
| "grad_norm": 2.9794514179229736, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.1906, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.88659793814433, | |
| "grad_norm": 2.7878856658935547, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 1.0781, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.9896907216494846, | |
| "eval_accuracy": 0.7573363431151241, | |
| "eval_loss": 0.9210801720619202, | |
| "eval_runtime": 8.2382, | |
| "eval_samples_per_second": 107.548, | |
| "eval_steps_per_second": 3.399, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.2989690721649483, | |
| "grad_norm": 3.0810065269470215, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.8921, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.711340206185567, | |
| "grad_norm": 3.1434895992279053, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.8049, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8036117381489842, | |
| "eval_loss": 0.7273964881896973, | |
| "eval_runtime": 8.2613, | |
| "eval_samples_per_second": 107.248, | |
| "eval_steps_per_second": 3.389, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 4.123711340206185, | |
| "grad_norm": 2.860553026199341, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.706, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.536082474226804, | |
| "grad_norm": 2.938096761703491, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.6356, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "grad_norm": 3.2027766704559326, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6091, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.989690721649485, | |
| "eval_accuracy": 0.8329571106094809, | |
| "eval_loss": 0.6427145004272461, | |
| "eval_runtime": 8.1064, | |
| "eval_samples_per_second": 109.297, | |
| "eval_steps_per_second": 3.454, | |
| "step": 242 | |
| }, | |
| { | |
| "epoch": 5.360824742268041, | |
| "grad_norm": 2.9634206295013428, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.504, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.77319587628866, | |
| "grad_norm": 2.7470180988311768, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.4985, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.8510158013544018, | |
| "eval_loss": 0.5518527626991272, | |
| "eval_runtime": 8.0512, | |
| "eval_samples_per_second": 110.045, | |
| "eval_steps_per_second": 3.478, | |
| "step": 291 | |
| }, | |
| { | |
| "epoch": 6.185567010309279, | |
| "grad_norm": 2.4766085147857666, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.4745, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.597938144329897, | |
| "grad_norm": 2.305265188217163, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.4077, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.989690721649485, | |
| "eval_accuracy": 0.8792325056433409, | |
| "eval_loss": 0.4920809268951416, | |
| "eval_runtime": 8.2019, | |
| "eval_samples_per_second": 108.024, | |
| "eval_steps_per_second": 3.414, | |
| "step": 339 | |
| }, | |
| { | |
| "epoch": 7.010309278350515, | |
| "grad_norm": 2.5866684913635254, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.3937, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.422680412371134, | |
| "grad_norm": 2.4354708194732666, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.3933, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.835051546391752, | |
| "grad_norm": 2.5438692569732666, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.3583, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8826185101580135, | |
| "eval_loss": 0.4755937159061432, | |
| "eval_runtime": 8.2595, | |
| "eval_samples_per_second": 107.271, | |
| "eval_steps_per_second": 3.39, | |
| "step": 388 | |
| }, | |
| { | |
| "epoch": 8.24742268041237, | |
| "grad_norm": 2.453233480453491, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.3537, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.65979381443299, | |
| "grad_norm": 2.6440579891204834, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.3292, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.989690721649485, | |
| "eval_accuracy": 0.8882618510158014, | |
| "eval_loss": 0.44721925258636475, | |
| "eval_runtime": 8.1718, | |
| "eval_samples_per_second": 108.422, | |
| "eval_steps_per_second": 3.426, | |
| "step": 436 | |
| }, | |
| { | |
| "epoch": 9.072164948453608, | |
| "grad_norm": 2.324805974960327, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.3014, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 9.484536082474227, | |
| "grad_norm": 2.4339663982391357, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.3151, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 9.896907216494846, | |
| "grad_norm": 2.0026326179504395, | |
| "learning_rate": 0.0, | |
| "loss": 0.338, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.896907216494846, | |
| "eval_accuracy": 0.8905191873589164, | |
| "eval_loss": 0.43482092022895813, | |
| "eval_runtime": 8.4069, | |
| "eval_samples_per_second": 105.389, | |
| "eval_steps_per_second": 3.331, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.896907216494846, | |
| "step": 480, | |
| "total_flos": 4.758713936025993e+18, | |
| "train_loss": 0.9664310624202093, | |
| "train_runtime": 1226.3415, | |
| "train_samples_per_second": 50.573, | |
| "train_steps_per_second": 0.391 | |
| } | |
| ], | |
| "logging_steps": 20, | |
| "max_steps": 480, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 4.758713936025993e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |