wav2vec2-base-EMOPIA / trainer_state.json
kurosekurose's picture
End of training
df58af8
raw
history blame
6.07 kB
{
"best_metric": 0.9441239833831787,
"best_model_checkpoint": "kurosekurose/wav2vec2-base-EMOPIA/checkpoint-525",
"epoch": 15.0,
"global_step": 2625,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 3.5e-06,
"loss": 1.3315,
"step": 175
},
{
"epoch": 1.0,
"eval_accuracy": 0.3857142925262451,
"eval_loss": 1.1984184980392456,
"eval_runtime": 48.1732,
"eval_samples_per_second": 1.453,
"eval_steps_per_second": 0.374,
"step": 175
},
{
"epoch": 2.0,
"learning_rate": 7e-06,
"loss": 1.0391,
"step": 350
},
{
"epoch": 2.0,
"eval_accuracy": 0.5428571701049805,
"eval_loss": 1.0071275234222412,
"eval_runtime": 48.3665,
"eval_samples_per_second": 1.447,
"eval_steps_per_second": 0.372,
"step": 350
},
{
"epoch": 3.0,
"learning_rate": 9.887058823529413e-06,
"loss": 0.8973,
"step": 525
},
{
"epoch": 3.0,
"eval_accuracy": 0.5857142806053162,
"eval_loss": 0.9441239833831787,
"eval_runtime": 48.2354,
"eval_samples_per_second": 1.451,
"eval_steps_per_second": 0.373,
"step": 525
},
{
"epoch": 4.0,
"learning_rate": 9.063529411764706e-06,
"loss": 0.8199,
"step": 700
},
{
"epoch": 4.0,
"eval_accuracy": 0.5285714268684387,
"eval_loss": 0.9983876347541809,
"eval_runtime": 48.536,
"eval_samples_per_second": 1.442,
"eval_steps_per_second": 0.371,
"step": 700
},
{
"epoch": 5.0,
"learning_rate": 8.244705882352942e-06,
"loss": 0.7675,
"step": 875
},
{
"epoch": 5.0,
"eval_accuracy": 0.6285714507102966,
"eval_loss": 0.9477005004882812,
"eval_runtime": 48.23,
"eval_samples_per_second": 1.451,
"eval_steps_per_second": 0.373,
"step": 875
},
{
"epoch": 6.0,
"learning_rate": 7.421176470588236e-06,
"loss": 0.6446,
"step": 1050
},
{
"epoch": 6.0,
"eval_accuracy": 0.6428571343421936,
"eval_loss": 1.0956463813781738,
"eval_runtime": 48.136,
"eval_samples_per_second": 1.454,
"eval_steps_per_second": 0.374,
"step": 1050
},
{
"epoch": 7.0,
"learning_rate": 6.59764705882353e-06,
"loss": 0.6221,
"step": 1225
},
{
"epoch": 7.0,
"eval_accuracy": 0.6285714507102966,
"eval_loss": 1.0143612623214722,
"eval_runtime": 48.3709,
"eval_samples_per_second": 1.447,
"eval_steps_per_second": 0.372,
"step": 1225
},
{
"epoch": 8.0,
"learning_rate": 5.778823529411765e-06,
"loss": 0.5543,
"step": 1400
},
{
"epoch": 8.0,
"eval_accuracy": 0.6285714507102966,
"eval_loss": 1.24734365940094,
"eval_runtime": 48.144,
"eval_samples_per_second": 1.454,
"eval_steps_per_second": 0.374,
"step": 1400
},
{
"epoch": 9.0,
"learning_rate": 4.955294117647059e-06,
"loss": 0.5095,
"step": 1575
},
{
"epoch": 9.0,
"eval_accuracy": 0.6714285612106323,
"eval_loss": 1.2690364122390747,
"eval_runtime": 48.163,
"eval_samples_per_second": 1.453,
"eval_steps_per_second": 0.374,
"step": 1575
},
{
"epoch": 10.0,
"learning_rate": 4.1317647058823535e-06,
"loss": 0.4461,
"step": 1750
},
{
"epoch": 10.0,
"eval_accuracy": 0.7285714149475098,
"eval_loss": 1.1364508867263794,
"eval_runtime": 48.2055,
"eval_samples_per_second": 1.452,
"eval_steps_per_second": 0.373,
"step": 1750
},
{
"epoch": 11.0,
"learning_rate": 3.3129411764705886e-06,
"loss": 0.3678,
"step": 1925
},
{
"epoch": 11.0,
"eval_accuracy": 0.699999988079071,
"eval_loss": 1.2065324783325195,
"eval_runtime": 48.5129,
"eval_samples_per_second": 1.443,
"eval_steps_per_second": 0.371,
"step": 1925
},
{
"epoch": 12.0,
"learning_rate": 2.4894117647058827e-06,
"loss": 0.3184,
"step": 2100
},
{
"epoch": 12.0,
"eval_accuracy": 0.6285714507102966,
"eval_loss": 1.699022889137268,
"eval_runtime": 48.0723,
"eval_samples_per_second": 1.456,
"eval_steps_per_second": 0.374,
"step": 2100
},
{
"epoch": 13.0,
"learning_rate": 1.6658823529411766e-06,
"loss": 0.271,
"step": 2275
},
{
"epoch": 13.0,
"eval_accuracy": 0.6285714507102966,
"eval_loss": 1.5335015058517456,
"eval_runtime": 48.3723,
"eval_samples_per_second": 1.447,
"eval_steps_per_second": 0.372,
"step": 2275
},
{
"epoch": 14.0,
"learning_rate": 8.423529411764707e-07,
"loss": 0.234,
"step": 2450
},
{
"epoch": 14.0,
"eval_accuracy": 0.7142857313156128,
"eval_loss": 1.3233639001846313,
"eval_runtime": 47.9369,
"eval_samples_per_second": 1.46,
"eval_steps_per_second": 0.375,
"step": 2450
},
{
"epoch": 15.0,
"learning_rate": 1.8823529411764708e-08,
"loss": 0.2049,
"step": 2625
},
{
"epoch": 15.0,
"eval_accuracy": 0.6428571343421936,
"eval_loss": 1.4966269731521606,
"eval_runtime": 48.2524,
"eval_samples_per_second": 1.451,
"eval_steps_per_second": 0.373,
"step": 2625
},
{
"epoch": 15.0,
"step": 2625,
"total_flos": 4.992891844826972e+18,
"train_loss": 0.6018750116257441,
"train_runtime": 20482.8582,
"train_samples_per_second": 0.513,
"train_steps_per_second": 0.128
}
],
"max_steps": 2625,
"num_train_epochs": 15,
"total_flos": 4.992891844826972e+18,
"trial_name": null,
"trial_params": null
}