musicgen-melody-lora-dnb-colab-8 / trainer_state.json
istvanj's picture
End of training
0208d05 verified
raw
history blame
5.53 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.0,
"eval_steps": 500,
"global_step": 56,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.26666666666666666,
"grad_norm": 13.590277671813965,
"learning_rate": 9.642857142857143e-05,
"loss": 9.0905,
"step": 2
},
{
"epoch": 0.5333333333333333,
"grad_norm": 17.921642303466797,
"learning_rate": 9.285714285714286e-05,
"loss": 8.6143,
"step": 4
},
{
"epoch": 0.8,
"grad_norm": 16.579307556152344,
"learning_rate": 8.92857142857143e-05,
"loss": 8.2238,
"step": 6
},
{
"epoch": 1.0,
"grad_norm": 5.841476917266846,
"learning_rate": 8.571428571428571e-05,
"loss": 5.9611,
"step": 8
},
{
"epoch": 1.2666666666666666,
"grad_norm": 13.555994033813477,
"learning_rate": 8.214285714285714e-05,
"loss": 7.7487,
"step": 10
},
{
"epoch": 1.5333333333333332,
"grad_norm": 13.151310920715332,
"learning_rate": 7.857142857142858e-05,
"loss": 7.5518,
"step": 12
},
{
"epoch": 1.8,
"grad_norm": 9.22488784790039,
"learning_rate": 7.500000000000001e-05,
"loss": 7.4163,
"step": 14
},
{
"epoch": 2.0,
"grad_norm": 4.5742621421813965,
"learning_rate": 7.142857142857143e-05,
"loss": 5.6047,
"step": 16
},
{
"epoch": 2.2666666666666666,
"grad_norm": 9.562870025634766,
"learning_rate": 6.785714285714286e-05,
"loss": 7.3735,
"step": 18
},
{
"epoch": 2.533333333333333,
"grad_norm": 6.718127250671387,
"learning_rate": 6.428571428571429e-05,
"loss": 7.3059,
"step": 20
},
{
"epoch": 2.8,
"grad_norm": 6.355175971984863,
"learning_rate": 6.0714285714285715e-05,
"loss": 7.3053,
"step": 22
},
{
"epoch": 3.0,
"grad_norm": 3.8696775436401367,
"learning_rate": 5.714285714285714e-05,
"loss": 5.4469,
"step": 24
},
{
"epoch": 3.2666666666666666,
"grad_norm": 6.373976707458496,
"learning_rate": 5.3571428571428575e-05,
"loss": 7.2302,
"step": 26
},
{
"epoch": 3.533333333333333,
"grad_norm": 6.881743907928467,
"learning_rate": 5e-05,
"loss": 7.2709,
"step": 28
},
{
"epoch": 3.8,
"grad_norm": 5.140980243682861,
"learning_rate": 4.642857142857143e-05,
"loss": 7.2163,
"step": 30
},
{
"epoch": 4.0,
"grad_norm": 2.9777579307556152,
"learning_rate": 4.2857142857142856e-05,
"loss": 5.3782,
"step": 32
},
{
"epoch": 4.266666666666667,
"grad_norm": 5.195846080780029,
"learning_rate": 4.107142857142857e-05,
"loss": 7.2058,
"step": 34
},
{
"epoch": 4.533333333333333,
"grad_norm": 7.618205547332764,
"learning_rate": 3.7500000000000003e-05,
"loss": 7.2148,
"step": 36
},
{
"epoch": 4.8,
"grad_norm": 4.3025665283203125,
"learning_rate": 3.392857142857143e-05,
"loss": 7.1619,
"step": 38
},
{
"epoch": 5.0,
"grad_norm": 3.16835880279541,
"learning_rate": 3.0357142857142857e-05,
"loss": 5.3649,
"step": 40
},
{
"epoch": 5.266666666666667,
"grad_norm": 6.764051914215088,
"learning_rate": 2.6785714285714288e-05,
"loss": 7.1409,
"step": 42
},
{
"epoch": 5.533333333333333,
"grad_norm": 3.742156744003296,
"learning_rate": 2.3214285714285715e-05,
"loss": 7.2022,
"step": 44
},
{
"epoch": 5.8,
"grad_norm": 7.770607948303223,
"learning_rate": 1.9642857142857145e-05,
"loss": 7.109,
"step": 46
},
{
"epoch": 6.0,
"grad_norm": 2.478787660598755,
"learning_rate": 1.6071428571428572e-05,
"loss": 5.3984,
"step": 48
},
{
"epoch": 6.266666666666667,
"grad_norm": 7.746104717254639,
"learning_rate": 1.25e-05,
"loss": 7.1566,
"step": 50
},
{
"epoch": 6.533333333333333,
"grad_norm": 4.224415302276611,
"learning_rate": 8.92857142857143e-06,
"loss": 7.162,
"step": 52
},
{
"epoch": 6.8,
"grad_norm": 3.788506269454956,
"learning_rate": 5.357142857142857e-06,
"loss": 7.0908,
"step": 54
},
{
"epoch": 7.0,
"grad_norm": 4.815220355987549,
"learning_rate": 1.7857142857142857e-06,
"loss": 5.3965,
"step": 56
},
{
"epoch": 7.0,
"step": 56,
"total_flos": 216424602812184.0,
"train_loss": 6.976511648723057,
"train_runtime": 1595.9192,
"train_samples_per_second": 0.602,
"train_steps_per_second": 0.035
}
],
"logging_steps": 2,
"max_steps": 56,
"num_input_tokens_seen": 0,
"num_train_epochs": 8,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 216424602812184.0,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}