JackWong0911's picture
End of training
c332aa0 verified
raw
history blame
13.2 kB
{
"best_metric": 0.96,
"best_model_checkpoint": "timesformer-base-finetuned-k400-kinetic400-subset-epoch6-num_frame_10_myViT2window3_more_data_b4/checkpoint-39",
"epoch": 5.097222222222222,
"eval_steps": 500,
"global_step": 72,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"accuracy": 0.0,
"epoch": 0,
"step": 0
},
{
"accuracy": 0.25,
"epoch": 0.01,
"step": 1
},
{
"accuracy": 0.25,
"epoch": 0.03,
"step": 2
},
{
"accuracy": 0.75,
"epoch": 0.04,
"step": 3
},
{
"accuracy": 0.0,
"epoch": 0.06,
"step": 4
},
{
"accuracy": 0.0,
"epoch": 0.07,
"step": 5
},
{
"accuracy": 0.5,
"epoch": 0.08,
"step": 6
},
{
"accuracy": 0.75,
"epoch": 0.1,
"step": 7
},
{
"accuracy": 0.75,
"epoch": 0.11,
"step": 8
},
{
"accuracy": 1.0,
"epoch": 0.12,
"step": 9
},
{
"epoch": 0.14,
"grad_norm": 11.168214797973633,
"learning_rate": 4.8437500000000005e-05,
"loss": 1.4033,
"step": 10
},
{
"accuracy": 0.75,
"epoch": 0.14,
"step": 10
},
{
"accuracy": 1.0,
"epoch": 0.15,
"step": 11
},
{
"accuracy": 1.0,
"epoch": 0.17,
"step": 12
},
{
"accuracy": 0.5,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 0.75,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 0.18,
"step": 13
},
{
"epoch": 0.18,
"eval_accuracy": 0.88,
"eval_loss": 0.5778742432594299,
"eval_runtime": 27.5022,
"eval_samples_per_second": 0.909,
"eval_steps_per_second": 0.255,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 0.18,
"step": 13
},
{
"accuracy": 1.0,
"epoch": 1.01,
"step": 14
},
{
"accuracy": 1.0,
"epoch": 1.03,
"step": 15
},
{
"accuracy": 1.0,
"epoch": 1.04,
"step": 16
},
{
"accuracy": 1.0,
"epoch": 1.06,
"step": 17
},
{
"accuracy": 1.0,
"epoch": 1.07,
"step": 18
},
{
"accuracy": 1.0,
"epoch": 1.08,
"step": 19
},
{
"epoch": 1.1,
"grad_norm": 2.1656227111816406,
"learning_rate": 4.0625000000000005e-05,
"loss": 0.3402,
"step": 20
},
{
"accuracy": 1.0,
"epoch": 1.1,
"step": 20
},
{
"accuracy": 1.0,
"epoch": 1.11,
"step": 21
},
{
"accuracy": 1.0,
"epoch": 1.12,
"step": 22
},
{
"accuracy": 1.0,
"epoch": 1.14,
"step": 23
},
{
"accuracy": 1.0,
"epoch": 1.15,
"step": 24
},
{
"accuracy": 1.0,
"epoch": 1.17,
"step": 25
},
{
"accuracy": 0.75,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 1.0,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 0.75,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 1.0,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 0.75,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 1.0,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 0.0,
"epoch": 1.18,
"step": 26
},
{
"epoch": 1.18,
"eval_accuracy": 0.84,
"eval_loss": 0.33898457884788513,
"eval_runtime": 27.8625,
"eval_samples_per_second": 0.897,
"eval_steps_per_second": 0.251,
"step": 26
},
{
"accuracy": 1.0,
"epoch": 1.18,
"step": 26
},
{
"accuracy": 1.0,
"epoch": 2.01,
"step": 27
},
{
"accuracy": 1.0,
"epoch": 2.03,
"step": 28
},
{
"accuracy": 1.0,
"epoch": 2.04,
"step": 29
},
{
"epoch": 2.06,
"grad_norm": 0.17987509071826935,
"learning_rate": 3.2812500000000005e-05,
"loss": 0.0426,
"step": 30
},
{
"accuracy": 1.0,
"epoch": 2.06,
"step": 30
},
{
"accuracy": 1.0,
"epoch": 2.07,
"step": 31
},
{
"accuracy": 1.0,
"epoch": 2.08,
"step": 32
},
{
"accuracy": 1.0,
"epoch": 2.1,
"step": 33
},
{
"accuracy": 1.0,
"epoch": 2.11,
"step": 34
},
{
"accuracy": 1.0,
"epoch": 2.12,
"step": 35
},
{
"accuracy": 1.0,
"epoch": 2.14,
"step": 36
},
{
"accuracy": 1.0,
"epoch": 2.15,
"step": 37
},
{
"accuracy": 1.0,
"epoch": 2.17,
"step": 38
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"accuracy": 0.75,
"epoch": 2.18,
"step": 39
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"epoch": 2.18,
"eval_accuracy": 0.96,
"eval_loss": 0.21377582848072052,
"eval_runtime": 27.0552,
"eval_samples_per_second": 0.924,
"eval_steps_per_second": 0.259,
"step": 39
},
{
"accuracy": 1.0,
"epoch": 2.18,
"step": 39
},
{
"epoch": 3.01,
"grad_norm": 0.11805030703544617,
"learning_rate": 2.5e-05,
"loss": 0.0121,
"step": 40
},
{
"accuracy": 1.0,
"epoch": 3.01,
"step": 40
},
{
"accuracy": 1.0,
"epoch": 3.03,
"step": 41
},
{
"accuracy": 1.0,
"epoch": 3.04,
"step": 42
},
{
"accuracy": 1.0,
"epoch": 3.06,
"step": 43
},
{
"accuracy": 1.0,
"epoch": 3.07,
"step": 44
},
{
"accuracy": 1.0,
"epoch": 3.08,
"step": 45
},
{
"accuracy": 1.0,
"epoch": 3.1,
"step": 46
},
{
"accuracy": 1.0,
"epoch": 3.11,
"step": 47
},
{
"accuracy": 1.0,
"epoch": 3.12,
"step": 48
},
{
"accuracy": 1.0,
"epoch": 3.14,
"step": 49
},
{
"epoch": 3.15,
"grad_norm": 0.07389847934246063,
"learning_rate": 1.71875e-05,
"loss": 0.0049,
"step": 50
},
{
"accuracy": 1.0,
"epoch": 3.15,
"step": 50
},
{
"accuracy": 1.0,
"epoch": 3.17,
"step": 51
},
{
"accuracy": 1.0,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 1.0,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 1.0,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 1.0,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 0.75,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 0.75,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 1.0,
"epoch": 3.18,
"step": 52
},
{
"epoch": 3.18,
"eval_accuracy": 0.92,
"eval_loss": 0.18079356849193573,
"eval_runtime": 27.6737,
"eval_samples_per_second": 0.903,
"eval_steps_per_second": 0.253,
"step": 52
},
{
"accuracy": 1.0,
"epoch": 3.18,
"step": 52
},
{
"accuracy": 1.0,
"epoch": 4.01,
"step": 53
},
{
"accuracy": 1.0,
"epoch": 4.03,
"step": 54
},
{
"accuracy": 1.0,
"epoch": 4.04,
"step": 55
},
{
"accuracy": 1.0,
"epoch": 4.06,
"step": 56
},
{
"accuracy": 1.0,
"epoch": 4.07,
"step": 57
},
{
"accuracy": 1.0,
"epoch": 4.08,
"step": 58
},
{
"accuracy": 1.0,
"epoch": 4.1,
"step": 59
},
{
"epoch": 4.11,
"grad_norm": 0.16879861056804657,
"learning_rate": 9.375000000000001e-06,
"loss": 0.0029,
"step": 60
},
{
"accuracy": 1.0,
"epoch": 4.11,
"step": 60
},
{
"accuracy": 1.0,
"epoch": 4.12,
"step": 61
},
{
"accuracy": 1.0,
"epoch": 4.14,
"step": 62
},
{
"accuracy": 1.0,
"epoch": 4.15,
"step": 63
},
{
"accuracy": 1.0,
"epoch": 4.17,
"step": 64
},
{
"accuracy": 1.0,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 0.75,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 1.0,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 1.0,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 0.75,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 1.0,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 1.0,
"epoch": 4.18,
"step": 65
},
{
"epoch": 4.18,
"eval_accuracy": 0.92,
"eval_loss": 0.1889605075120926,
"eval_runtime": 25.2006,
"eval_samples_per_second": 0.992,
"eval_steps_per_second": 0.278,
"step": 65
},
{
"accuracy": 1.0,
"epoch": 4.18,
"step": 65
},
{
"accuracy": 1.0,
"epoch": 5.01,
"step": 66
},
{
"accuracy": 1.0,
"epoch": 5.03,
"step": 67
},
{
"accuracy": 1.0,
"epoch": 5.04,
"step": 68
},
{
"accuracy": 1.0,
"epoch": 5.06,
"step": 69
},
{
"epoch": 5.07,
"grad_norm": 0.03316669911146164,
"learning_rate": 1.5625e-06,
"loss": 0.0022,
"step": 70
},
{
"accuracy": 1.0,
"epoch": 5.07,
"step": 70
},
{
"accuracy": 1.0,
"epoch": 5.08,
"step": 71
},
{
"accuracy": 0.75,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 0.75,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"epoch": 5.1,
"eval_accuracy": 0.92,
"eval_loss": 0.18929095566272736,
"eval_runtime": 25.0662,
"eval_samples_per_second": 0.997,
"eval_steps_per_second": 0.279,
"step": 72
},
{
"epoch": 5.1,
"step": 72,
"total_flos": 3.044708388358963e+17,
"train_loss": 0.25122544505090144,
"train_runtime": 480.2289,
"train_samples_per_second": 0.6,
"train_steps_per_second": 0.15
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 0.0,
"epoch": 5.1,
"step": 72
},
{
"epoch": 5.1,
"eval_accuracy": 0.96,
"eval_loss": 0.1763344556093216,
"eval_runtime": 17.7295,
"eval_samples_per_second": 1.41,
"eval_steps_per_second": 0.395,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 0.75,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"accuracy": 1.0,
"epoch": 5.1,
"step": 72
},
{
"epoch": 5.1,
"eval_accuracy": 0.96,
"eval_loss": 0.1763344407081604,
"eval_runtime": 17.9669,
"eval_samples_per_second": 1.391,
"eval_steps_per_second": 0.39,
"step": 72
}
],
"logging_steps": 10,
"max_steps": 72,
"num_input_tokens_seen": 0,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 3.044708388358963e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}