longformer-spans / checkpoint-567 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 7, checkpoint
206120c verified
raw
history blame
8.48 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 7.0,
"eval_steps": 500,
"global_step": 567,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_B": {
"f1-score": 0.7905866302864938,
"precision": 0.7768096514745308,
"recall": 0.8048611111111111,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9005331227169513,
"precision": 0.9637593111099372,
"recall": 0.8450919534905267,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.8288892663043479,
"precision": 0.7463108800367001,
"recall": 0.9320156593144275,
"support": 10473.0
},
"eval_accuracy": 0.8705373134328358,
"eval_loss": 0.32930415868759155,
"eval_macro avg": {
"f1-score": 0.8400030064359308,
"precision": 0.8289599475403894,
"recall": 0.8606562413053552,
"support": 33500.0
},
"eval_runtime": 4.7349,
"eval_samples_per_second": 17.107,
"eval_steps_per_second": 2.323,
"eval_weighted avg": {
"f1-score": 0.8734092702599646,
"precision": 0.8877430445874,
"recall": 0.8705373134328358,
"support": 33500.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_B": {
"f1-score": 0.8656218739579861,
"precision": 0.8325849903784477,
"recall": 0.9013888888888889,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9547168948727912,
"precision": 0.9423736462093862,
"recall": 0.9673877796822161,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.9013528191962081,
"precision": 0.9332379102341274,
"recall": 0.8715745249689678,
"support": 10473.0
},
"eval_accuracy": 0.9345970149253732,
"eval_loss": 0.2162298709154129,
"eval_macro avg": {
"f1-score": 0.9072305293423284,
"precision": 0.902732182273987,
"recall": 0.9134503978466909,
"support": 33500.0
},
"eval_runtime": 4.753,
"eval_samples_per_second": 17.042,
"eval_steps_per_second": 2.314,
"eval_weighted avg": {
"f1-score": 0.9342040950316517,
"precision": 0.9347982961417612,
"recall": 0.9345970149253732,
"support": 33500.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_B": {
"f1-score": 0.8603587759409076,
"precision": 0.8717034925160371,
"recall": 0.8493055555555555,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9434195218082959,
"precision": 0.9677057963955188,
"recall": 0.9203224162690509,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.8879310344827586,
"precision": 0.8459410391631365,
"recall": 0.9343072663038289,
"support": 10473.0
},
"eval_accuracy": 0.9216417910447762,
"eval_loss": 0.24082203209400177,
"eval_macro avg": {
"f1-score": 0.8972364440773206,
"precision": 0.8951167760248975,
"recall": 0.9013117460428117,
"support": 33500.0
},
"eval_runtime": 4.7554,
"eval_samples_per_second": 17.033,
"eval_steps_per_second": 2.313,
"eval_weighted avg": {
"f1-score": 0.9225019575751797,
"precision": 0.9255121957960801,
"recall": 0.9216417910447762,
"support": 33500.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_B": {
"f1-score": 0.8804020100502512,
"precision": 0.8504854368932039,
"recall": 0.9125,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9589184992350025,
"precision": 0.945595388218339,
"recall": 0.9726224116366332,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.9091178797468354,
"precision": 0.9427750999897446,
"recall": 0.877780960565263,
"support": 10473.0
},
"eval_accuracy": 0.9403880597014925,
"eval_loss": 0.22083505988121033,
"eval_macro avg": {
"f1-score": 0.9161461296773631,
"precision": 0.9129519750337626,
"recall": 0.9209677907339654,
"support": 33500.0
},
"eval_runtime": 4.7449,
"eval_samples_per_second": 17.071,
"eval_steps_per_second": 2.318,
"eval_weighted avg": {
"f1-score": 0.9399744505088647,
"precision": 0.9406253819936745,
"recall": 0.9403880597014925,
"support": 33500.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_B": {
"f1-score": 0.8879781420765027,
"precision": 0.8736559139784946,
"recall": 0.9027777777777778,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9613201259451634,
"precision": 0.9539317642765919,
"recall": 0.9688238291564367,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.916006030835076,
"precision": 0.9334853291038858,
"recall": 0.899169292466342,
"support": 10473.0
},
"eval_accuracy": 0.9442089552238806,
"eval_loss": 0.2257838249206543,
"eval_macro avg": {
"f1-score": 0.921768099618914,
"precision": 0.9203576691196576,
"recall": 0.9235902998001855,
"support": 33500.0
},
"eval_runtime": 4.7845,
"eval_samples_per_second": 16.93,
"eval_steps_per_second": 2.299,
"eval_weighted avg": {
"f1-score": 0.9440011117702735,
"precision": 0.9440889959185916,
"recall": 0.9442089552238806,
"support": 33500.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_B": {
"f1-score": 0.8873579056148813,
"precision": 0.8803827751196173,
"recall": 0.8944444444444445,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9622060984570168,
"precision": 0.9539267015706806,
"recall": 0.9706304720433594,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.9182769530299342,
"precision": 0.9365567911040509,
"recall": 0.9006970304592762,
"support": 10473.0
},
"eval_accuracy": 0.9454925373134329,
"eval_loss": 0.2242102175951004,
"eval_macro avg": {
"f1-score": 0.9226136523672773,
"precision": 0.923622089264783,
"recall": 0.9219239823156933,
"support": 33500.0
},
"eval_runtime": 4.7715,
"eval_samples_per_second": 16.976,
"eval_steps_per_second": 2.305,
"eval_weighted avg": {
"f1-score": 0.945255312255509,
"precision": 0.9453351097376493,
"recall": 0.9454925373134329,
"support": 33500.0
},
"step": 486
},
{
"epoch": 6.17,
"grad_norm": 1.8705081939697266,
"learning_rate": 1.7530864197530865e-05,
"loss": 0.1601,
"step": 500
},
{
"epoch": 7.0,
"eval_B": {
"f1-score": 0.8876749743939911,
"precision": 0.8730691739422431,
"recall": 0.9027777777777778,
"support": 1440.0
},
"eval_I": {
"f1-score": 0.9611415113943341,
"precision": 0.9581108451482232,
"recall": 0.9641914114976606,
"support": 21587.0
},
"eval_O": {
"f1-score": 0.9164739884393064,
"precision": 0.9247594050743657,
"recall": 0.9083357204239473,
"support": 10473.0
},
"eval_accuracy": 0.944089552238806,
"eval_loss": 0.22517083585262299,
"eval_macro avg": {
"f1-score": 0.9217634914092105,
"precision": 0.9186464747216107,
"recall": 0.9251016365664619,
"support": 33500.0
},
"eval_runtime": 4.7926,
"eval_samples_per_second": 16.901,
"eval_steps_per_second": 2.295,
"eval_weighted avg": {
"f1-score": 0.9440192791200506,
"precision": 0.9440287663891153,
"recall": 0.944089552238806,
"support": 33500.0
},
"step": 567
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 1003380772350600.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}