MattyB95's picture
End of training
11c0b65
raw
history blame
4.2 kB
{
"best_metric": 0.002155957045033574,
"best_model_checkpoint": "W:/res/Transformers/my-google/vit-base-patch16-224-in21k\\checkpoint-6346",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 9519,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.16,
"learning_rate": 4.7373673705221136e-05,
"loss": 0.0936,
"step": 500
},
{
"epoch": 0.32,
"learning_rate": 4.4747347410442276e-05,
"loss": 0.019,
"step": 1000
},
{
"epoch": 0.47,
"learning_rate": 4.212102111566341e-05,
"loss": 0.0339,
"step": 1500
},
{
"epoch": 0.63,
"learning_rate": 3.949469482088455e-05,
"loss": 0.0129,
"step": 2000
},
{
"epoch": 0.79,
"learning_rate": 3.686836852610568e-05,
"loss": 0.0154,
"step": 2500
},
{
"epoch": 0.95,
"learning_rate": 3.424204223132682e-05,
"loss": 0.007,
"step": 3000
},
{
"epoch": 1.0,
"eval_accuracy": 0.9971824182901304,
"eval_f1": 0.9984326720702164,
"eval_loss": 0.010775323025882244,
"eval_precision": 0.9968702494858267,
"eval_recall": 1.0,
"eval_runtime": 484.8551,
"eval_samples_per_second": 51.24,
"eval_steps_per_second": 6.406,
"step": 3173
},
{
"epoch": 1.1,
"learning_rate": 3.1615715936547956e-05,
"loss": 0.0068,
"step": 3500
},
{
"epoch": 1.26,
"learning_rate": 2.8989389641769092e-05,
"loss": 0.0047,
"step": 4000
},
{
"epoch": 1.42,
"learning_rate": 2.636306334699023e-05,
"loss": 0.0041,
"step": 4500
},
{
"epoch": 1.58,
"learning_rate": 2.373673705221137e-05,
"loss": 0.0022,
"step": 5000
},
{
"epoch": 1.73,
"learning_rate": 2.1110410757432505e-05,
"loss": 0.0028,
"step": 5500
},
{
"epoch": 1.89,
"learning_rate": 1.8484084462653642e-05,
"loss": 0.0015,
"step": 6000
},
{
"epoch": 2.0,
"eval_accuracy": 0.9996779906617292,
"eval_f1": 0.9998205875756896,
"eval_loss": 0.002155957045033574,
"eval_precision": 0.9998654346460931,
"eval_recall": 0.9997757445281665,
"eval_runtime": 380.0049,
"eval_samples_per_second": 65.378,
"eval_steps_per_second": 8.174,
"step": 6346
},
{
"epoch": 2.05,
"learning_rate": 1.585775816787478e-05,
"loss": 0.0001,
"step": 6500
},
{
"epoch": 2.21,
"learning_rate": 1.3231431873095915e-05,
"loss": 0.0001,
"step": 7000
},
{
"epoch": 2.36,
"learning_rate": 1.060510557831705e-05,
"loss": 0.003,
"step": 7500
},
{
"epoch": 2.52,
"learning_rate": 7.978779283538187e-06,
"loss": 0.0001,
"step": 8000
},
{
"epoch": 2.68,
"learning_rate": 5.352452988759324e-06,
"loss": 0.0,
"step": 8500
},
{
"epoch": 2.84,
"learning_rate": 2.7261266939804603e-06,
"loss": 0.0,
"step": 9000
},
{
"epoch": 2.99,
"learning_rate": 9.98003992015968e-08,
"loss": 0.0,
"step": 9500
},
{
"epoch": 3.0,
"eval_accuracy": 0.9996377394944453,
"eval_f1": 0.9997981836528759,
"eval_loss": 0.0025263545103371143,
"eval_precision": 0.9997309296381004,
"eval_recall": 0.9998654467168999,
"eval_runtime": 380.4752,
"eval_samples_per_second": 65.297,
"eval_steps_per_second": 8.163,
"step": 9519
},
{
"epoch": 3.0,
"step": 9519,
"total_flos": 5.900240089255035e+18,
"train_loss": 0.010893226607249389,
"train_runtime": 2856.1574,
"train_samples_per_second": 26.658,
"train_steps_per_second": 3.333
}
],
"logging_steps": 500,
"max_steps": 9519,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 5.900240089255035e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}