rvv-karma's picture
End of training
743363a
{
"best_metric": 0.8785714285714286,
"best_model_checkpoint": "Human-Action-Recognition-VIT-Base-patch16-224/checkpoint-316",
"epoch": 19.746835443037973,
"eval_steps": 500,
"global_step": 780,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.99,
"learning_rate": 2.5e-05,
"loss": 2.6396,
"step": 39
},
{
"epoch": 0.99,
"eval_accuracy": 0.44246031746031744,
"eval_loss": 2.04364275932312,
"eval_runtime": 33.6234,
"eval_samples_per_second": 74.948,
"eval_steps_per_second": 1.19,
"step": 39
},
{
"epoch": 2.0,
"learning_rate": 4.992877492877493e-05,
"loss": 1.4579,
"step": 79
},
{
"epoch": 2.0,
"eval_accuracy": 0.7916666666666666,
"eval_loss": 0.7552899122238159,
"eval_runtime": 33.9732,
"eval_samples_per_second": 74.176,
"eval_steps_per_second": 1.177,
"step": 79
},
{
"epoch": 2.99,
"learning_rate": 4.7150997150997157e-05,
"loss": 0.8342,
"step": 118
},
{
"epoch": 2.99,
"eval_accuracy": 0.8416666666666667,
"eval_loss": 0.5296158194541931,
"eval_runtime": 33.8808,
"eval_samples_per_second": 74.378,
"eval_steps_per_second": 1.181,
"step": 118
},
{
"epoch": 4.0,
"learning_rate": 4.4301994301994304e-05,
"loss": 0.6649,
"step": 158
},
{
"epoch": 4.0,
"eval_accuracy": 0.8496031746031746,
"eval_loss": 0.4977969229221344,
"eval_runtime": 33.6309,
"eval_samples_per_second": 74.931,
"eval_steps_per_second": 1.189,
"step": 158
},
{
"epoch": 4.99,
"learning_rate": 4.152421652421652e-05,
"loss": 0.6137,
"step": 197
},
{
"epoch": 4.99,
"eval_accuracy": 0.8595238095238096,
"eval_loss": 0.44600579142570496,
"eval_runtime": 33.559,
"eval_samples_per_second": 75.092,
"eval_steps_per_second": 1.192,
"step": 197
},
{
"epoch": 6.0,
"learning_rate": 3.867521367521368e-05,
"loss": 0.5374,
"step": 237
},
{
"epoch": 6.0,
"eval_accuracy": 0.8626984126984127,
"eval_loss": 0.43557941913604736,
"eval_runtime": 33.4656,
"eval_samples_per_second": 75.301,
"eval_steps_per_second": 1.195,
"step": 237
},
{
"epoch": 6.99,
"learning_rate": 3.58974358974359e-05,
"loss": 0.514,
"step": 276
},
{
"epoch": 6.99,
"eval_accuracy": 0.8615079365079366,
"eval_loss": 0.4348864257335663,
"eval_runtime": 33.7031,
"eval_samples_per_second": 74.771,
"eval_steps_per_second": 1.187,
"step": 276
},
{
"epoch": 8.0,
"learning_rate": 3.304843304843305e-05,
"loss": 0.475,
"step": 316
},
{
"epoch": 8.0,
"eval_accuracy": 0.8785714285714286,
"eval_loss": 0.4005189836025238,
"eval_runtime": 33.4656,
"eval_samples_per_second": 75.301,
"eval_steps_per_second": 1.195,
"step": 316
},
{
"epoch": 8.99,
"learning_rate": 3.0270655270655275e-05,
"loss": 0.4663,
"step": 355
},
{
"epoch": 8.99,
"eval_accuracy": 0.8658730158730159,
"eval_loss": 0.4163670241832733,
"eval_runtime": 33.7779,
"eval_samples_per_second": 74.605,
"eval_steps_per_second": 1.184,
"step": 355
},
{
"epoch": 10.0,
"learning_rate": 2.7421652421652423e-05,
"loss": 0.4178,
"step": 395
},
{
"epoch": 10.0,
"eval_accuracy": 0.8738095238095238,
"eval_loss": 0.4127999544143677,
"eval_runtime": 33.3961,
"eval_samples_per_second": 75.458,
"eval_steps_per_second": 1.198,
"step": 395
},
{
"epoch": 10.99,
"learning_rate": 2.4643874643874645e-05,
"loss": 0.4226,
"step": 434
},
{
"epoch": 10.99,
"eval_accuracy": 0.8690476190476191,
"eval_loss": 0.4115118980407715,
"eval_runtime": 33.5412,
"eval_samples_per_second": 75.131,
"eval_steps_per_second": 1.193,
"step": 434
},
{
"epoch": 12.0,
"learning_rate": 2.1794871794871795e-05,
"loss": 0.3896,
"step": 474
},
{
"epoch": 12.0,
"eval_accuracy": 0.875,
"eval_loss": 0.4112005829811096,
"eval_runtime": 33.6831,
"eval_samples_per_second": 74.815,
"eval_steps_per_second": 1.188,
"step": 474
},
{
"epoch": 12.99,
"learning_rate": 1.9017094017094017e-05,
"loss": 0.3866,
"step": 513
},
{
"epoch": 12.99,
"eval_accuracy": 0.8714285714285714,
"eval_loss": 0.4072014093399048,
"eval_runtime": 33.7064,
"eval_samples_per_second": 74.763,
"eval_steps_per_second": 1.187,
"step": 513
},
{
"epoch": 14.0,
"learning_rate": 1.6168091168091168e-05,
"loss": 0.3632,
"step": 553
},
{
"epoch": 14.0,
"eval_accuracy": 0.8718253968253968,
"eval_loss": 0.41061681509017944,
"eval_runtime": 33.9136,
"eval_samples_per_second": 74.306,
"eval_steps_per_second": 1.179,
"step": 553
},
{
"epoch": 14.99,
"learning_rate": 1.3390313390313392e-05,
"loss": 0.3596,
"step": 592
},
{
"epoch": 14.99,
"eval_accuracy": 0.8714285714285714,
"eval_loss": 0.4042780101299286,
"eval_runtime": 33.971,
"eval_samples_per_second": 74.181,
"eval_steps_per_second": 1.177,
"step": 592
},
{
"epoch": 16.0,
"learning_rate": 1.0541310541310543e-05,
"loss": 0.3421,
"step": 632
},
{
"epoch": 16.0,
"eval_accuracy": 0.8674603174603175,
"eval_loss": 0.4128040075302124,
"eval_runtime": 33.5678,
"eval_samples_per_second": 75.072,
"eval_steps_per_second": 1.192,
"step": 632
},
{
"epoch": 16.99,
"learning_rate": 7.763532763532765e-06,
"loss": 0.344,
"step": 671
},
{
"epoch": 16.99,
"eval_accuracy": 0.8642857142857143,
"eval_loss": 0.41805535554885864,
"eval_runtime": 33.6623,
"eval_samples_per_second": 74.861,
"eval_steps_per_second": 1.188,
"step": 671
},
{
"epoch": 18.0,
"learning_rate": 4.914529914529915e-06,
"loss": 0.3447,
"step": 711
},
{
"epoch": 18.0,
"eval_accuracy": 0.8686507936507937,
"eval_loss": 0.41281768679618835,
"eval_runtime": 33.9679,
"eval_samples_per_second": 74.188,
"eval_steps_per_second": 1.178,
"step": 711
},
{
"epoch": 18.99,
"learning_rate": 2.136752136752137e-06,
"loss": 0.3407,
"step": 750
},
{
"epoch": 18.99,
"eval_accuracy": 0.8714285714285714,
"eval_loss": 0.40965768694877625,
"eval_runtime": 33.6119,
"eval_samples_per_second": 74.973,
"eval_steps_per_second": 1.19,
"step": 750
},
{
"epoch": 19.75,
"learning_rate": 0.0,
"loss": 0.3267,
"step": 780
},
{
"epoch": 19.75,
"eval_accuracy": 0.8682539682539683,
"eval_loss": 0.40967872738838196,
"eval_runtime": 34.1352,
"eval_samples_per_second": 73.824,
"eval_steps_per_second": 1.172,
"step": 780
},
{
"epoch": 19.75,
"step": 780,
"total_flos": 1.5428282771770638e+19,
"train_loss": 0.6146706715608254,
"train_runtime": 7303.7952,
"train_samples_per_second": 27.602,
"train_steps_per_second": 0.107
}
],
"logging_steps": 500,
"max_steps": 780,
"num_train_epochs": 20,
"save_steps": 500,
"total_flos": 1.5428282771770638e+19,
"trial_name": null,
"trial_params": null
}