vit-artworkclassifier / trainer_state.json
oschamp's picture
End of training
bc0d2d6
{
"best_metric": 1.1392158269882202,
"best_model_checkpoint": "./vit-artworkclassifier\\checkpoint-600",
"epoch": 4.0,
"global_step": 1112,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.18,
"learning_rate": 9.550359712230216e-05,
"loss": 1.9425,
"step": 50
},
{
"epoch": 0.36,
"learning_rate": 9.100719424460432e-05,
"loss": 1.5906,
"step": 100
},
{
"epoch": 0.36,
"eval_accuracy": 0.48467650397275824,
"eval_loss": 1.4708987474441528,
"eval_runtime": 7.3715,
"eval_samples_per_second": 119.514,
"eval_steps_per_second": 15.058,
"step": 100
},
{
"epoch": 0.54,
"learning_rate": 8.651079136690648e-05,
"loss": 1.4388,
"step": 150
},
{
"epoch": 0.72,
"learning_rate": 8.201438848920863e-05,
"loss": 1.3395,
"step": 200
},
{
"epoch": 0.72,
"eval_accuracy": 0.5073779795686719,
"eval_loss": 1.3207731246948242,
"eval_runtime": 7.4035,
"eval_samples_per_second": 118.998,
"eval_steps_per_second": 14.993,
"step": 200
},
{
"epoch": 0.9,
"learning_rate": 7.751798561151079e-05,
"loss": 1.3275,
"step": 250
},
{
"epoch": 1.08,
"learning_rate": 7.302158273381295e-05,
"loss": 1.1461,
"step": 300
},
{
"epoch": 1.08,
"eval_accuracy": 0.5164585698070374,
"eval_loss": 1.3362975120544434,
"eval_runtime": 7.64,
"eval_samples_per_second": 115.314,
"eval_steps_per_second": 14.529,
"step": 300
},
{
"epoch": 1.26,
"learning_rate": 6.852517985611511e-05,
"loss": 0.9749,
"step": 350
},
{
"epoch": 1.44,
"learning_rate": 6.402877697841726e-05,
"loss": 0.9593,
"step": 400
},
{
"epoch": 1.44,
"eval_accuracy": 0.5845629965947786,
"eval_loss": 1.1789547204971313,
"eval_runtime": 7.248,
"eval_samples_per_second": 121.551,
"eval_steps_per_second": 15.315,
"step": 400
},
{
"epoch": 1.62,
"learning_rate": 5.953237410071942e-05,
"loss": 0.9175,
"step": 450
},
{
"epoch": 1.8,
"learning_rate": 5.512589928057554e-05,
"loss": 0.8761,
"step": 500
},
{
"epoch": 1.8,
"eval_accuracy": 0.5902383654937571,
"eval_loss": 1.1252126693725586,
"eval_runtime": 7.2535,
"eval_samples_per_second": 121.459,
"eval_steps_per_second": 15.303,
"step": 500
},
{
"epoch": 1.98,
"learning_rate": 5.06294964028777e-05,
"loss": 0.8615,
"step": 550
},
{
"epoch": 2.16,
"learning_rate": 4.613309352517986e-05,
"loss": 0.5922,
"step": 600
},
{
"epoch": 2.16,
"eval_accuracy": 0.5947786606129398,
"eval_loss": 1.1392158269882202,
"eval_runtime": 7.5695,
"eval_samples_per_second": 116.388,
"eval_steps_per_second": 14.664,
"step": 600
},
{
"epoch": 2.34,
"learning_rate": 4.163669064748202e-05,
"loss": 0.4684,
"step": 650
},
{
"epoch": 2.52,
"learning_rate": 3.714028776978418e-05,
"loss": 0.4803,
"step": 700
},
{
"epoch": 2.52,
"eval_accuracy": 0.5936435868331441,
"eval_loss": 1.1560313701629639,
"eval_runtime": 7.3175,
"eval_samples_per_second": 120.396,
"eval_steps_per_second": 15.169,
"step": 700
},
{
"epoch": 2.7,
"learning_rate": 3.2643884892086335e-05,
"loss": 0.4766,
"step": 750
},
{
"epoch": 2.88,
"learning_rate": 2.8147482014388493e-05,
"loss": 0.4454,
"step": 800
},
{
"epoch": 2.88,
"eval_accuracy": 0.6118047673098751,
"eval_loss": 1.1545273065567017,
"eval_runtime": 7.2475,
"eval_samples_per_second": 121.559,
"eval_steps_per_second": 15.316,
"step": 800
},
{
"epoch": 3.06,
"learning_rate": 2.3651079136690647e-05,
"loss": 0.371,
"step": 850
},
{
"epoch": 3.24,
"learning_rate": 1.9154676258992808e-05,
"loss": 0.2271,
"step": 900
},
{
"epoch": 3.24,
"eval_accuracy": 0.6038592508513053,
"eval_loss": 1.2283731698989868,
"eval_runtime": 7.1705,
"eval_samples_per_second": 122.865,
"eval_steps_per_second": 15.48,
"step": 900
},
{
"epoch": 3.42,
"learning_rate": 1.4658273381294965e-05,
"loss": 0.2249,
"step": 950
},
{
"epoch": 3.6,
"learning_rate": 1.0161870503597123e-05,
"loss": 0.207,
"step": 1000
},
{
"epoch": 3.6,
"eval_accuracy": 0.5959137343927355,
"eval_loss": 1.262502670288086,
"eval_runtime": 7.2385,
"eval_samples_per_second": 121.71,
"eval_steps_per_second": 15.335,
"step": 1000
},
{
"epoch": 3.78,
"learning_rate": 5.665467625899281e-06,
"loss": 0.1906,
"step": 1050
},
{
"epoch": 3.96,
"learning_rate": 1.169064748201439e-06,
"loss": 0.1958,
"step": 1100
},
{
"epoch": 3.96,
"eval_accuracy": 0.6004540295119183,
"eval_loss": 1.262100338935852,
"eval_runtime": 7.2855,
"eval_samples_per_second": 120.925,
"eval_steps_per_second": 15.236,
"step": 1100
},
{
"epoch": 4.0,
"step": 1112,
"total_flos": 2.7542380976164454e+18,
"train_loss": 0.7775138853479633,
"train_runtime": 613.7183,
"train_samples_per_second": 57.909,
"train_steps_per_second": 1.812
}
],
"max_steps": 1112,
"num_train_epochs": 4,
"total_flos": 2.7542380976164454e+18,
"trial_name": null,
"trial_params": null
}