trainer_state.json

{
  "best_metric": 0.5063291139240507,
  "best_model_checkpoint": "videomae-base-finetuned-kinetics-finetuned-caer-final\\checkpoint-146",
  "epoch": 1.4383561643835616,
  "eval_steps": 500,
  "global_step": 146,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "grad_norm": 5.686412811279297,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.6544,
      "step": 10
    },
    {
      "epoch": 0.14,
      "grad_norm": 5.574044704437256,
      "learning_rate": 4.809160305343512e-05,
      "loss": 1.5654,
      "step": 20
    },
    {
      "epoch": 0.21,
      "grad_norm": 5.577368259429932,
      "learning_rate": 4.4274809160305345e-05,
      "loss": 1.5835,
      "step": 30
    },
    {
      "epoch": 0.27,
      "grad_norm": 6.698699474334717,
      "learning_rate": 4.0458015267175576e-05,
      "loss": 1.531,
      "step": 40
    },
    {
      "epoch": 0.34,
      "grad_norm": 7.928526878356934,
      "learning_rate": 3.66412213740458e-05,
      "loss": 1.5016,
      "step": 50
    },
    {
      "epoch": 0.41,
      "grad_norm": 9.12410831451416,
      "learning_rate": 3.282442748091603e-05,
      "loss": 1.3787,
      "step": 60
    },
    {
      "epoch": 0.48,
      "grad_norm": 7.112976551055908,
      "learning_rate": 2.900763358778626e-05,
      "loss": 1.48,
      "step": 70
    },
    {
      "epoch": 0.55,
      "grad_norm": 6.422354698181152,
      "learning_rate": 2.5190839694656487e-05,
      "loss": 1.4128,
      "step": 80
    },
    {
      "epoch": 0.56,
      "eval_accuracy": 0.35443037974683544,
      "eval_loss": 1.4815402030944824,
      "eval_runtime": 28.5428,
      "eval_samples_per_second": 2.768,
      "eval_steps_per_second": 0.49,
      "step": 82
    },
    {
      "epoch": 1.05,
      "grad_norm": 6.360805511474609,
      "learning_rate": 2.1374045801526718e-05,
      "loss": 1.3173,
      "step": 90
    },
    {
      "epoch": 1.12,
      "grad_norm": 8.944757461547852,
      "learning_rate": 1.7557251908396945e-05,
      "loss": 1.1789,
      "step": 100
    },
    {
      "epoch": 1.19,
      "grad_norm": 8.334917068481445,
      "learning_rate": 1.3740458015267178e-05,
      "loss": 1.2477,
      "step": 110
    },
    {
      "epoch": 1.26,
      "grad_norm": 11.71667194366455,
      "learning_rate": 9.923664122137405e-06,
      "loss": 1.1704,
      "step": 120
    },
    {
      "epoch": 1.33,
      "grad_norm": 11.985795021057129,
      "learning_rate": 6.106870229007634e-06,
      "loss": 1.0841,
      "step": 130
    },
    {
      "epoch": 1.4,
      "grad_norm": 7.384586334228516,
      "learning_rate": 2.2900763358778625e-06,
      "loss": 1.0255,
      "step": 140
    },
    {
      "epoch": 1.44,
      "eval_accuracy": 0.5063291139240507,
      "eval_loss": 1.3321611881256104,
      "eval_runtime": 28.5122,
      "eval_samples_per_second": 2.771,
      "eval_steps_per_second": 0.491,
      "step": 146
    },
    {
      "epoch": 1.44,
      "step": 146,
      "total_flos": 1.0865973925262131e+18,
      "train_loss": 1.3585324875295979,
      "train_runtime": 492.8068,
      "train_samples_per_second": 1.778,
      "train_steps_per_second": 0.296
    },
    {
      "epoch": 1.44,
      "eval_accuracy": 0.5,
      "eval_loss": 1.355814814567566,
      "eval_runtime": 18.5285,
      "eval_samples_per_second": 3.022,
      "eval_steps_per_second": 0.54,
      "step": 146
    },
    {
      "epoch": 1.44,
      "eval_accuracy": 0.5063291139240507,
      "eval_loss": 1.3321610689163208,
      "eval_runtime": 31.1651,
      "eval_samples_per_second": 2.535,
      "eval_steps_per_second": 0.449,
      "step": 146
    }
  ],
  "logging_steps": 10,
  "max_steps": 146,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9223372036854775807,
  "save_steps": 500,
  "total_flos": 1.0865973925262131e+18,
  "train_batch_size": 6,
  "trial_name": null,
  "trial_params": null
}