Bert_v12 / trainer_state.json
JhonMR's picture
End of training
a32d33b verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 9.0,
"eval_steps": 500,
"global_step": 2484,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"step": 276,
"train_eval_accuracy": 0.5786848072562358,
"train_eval_f1": 0.5273499989435928,
"train_eval_loss": 1.8481072187423706,
"train_eval_precision": 0.5941973878901876,
"train_eval_recall": 0.572395136508943,
"train_loss": 1.8481072187423706,
"train_runtime": 268.102,
"train_samples_per_second": 32.898,
"train_steps_per_second": 1.029
},
{
"epoch": 1.0,
"eval_accuracy": 0.5494708994708994,
"eval_f1": 0.5078140359989309,
"eval_loss": 1.8939365148544312,
"eval_precision": 0.5816333388253621,
"eval_recall": 0.5635339213221277,
"eval_runtime": 114.8487,
"eval_samples_per_second": 32.913,
"eval_steps_per_second": 1.036,
"step": 276
},
{
"epoch": 2.0,
"step": 552,
"train_eval_accuracy": 0.8439909297052154,
"train_eval_f1": 0.8342492924641463,
"train_eval_loss": 0.8016859889030457,
"train_eval_precision": 0.8542755916165877,
"train_eval_recall": 0.8407044685499563,
"train_loss": 0.8016859292984009,
"train_runtime": 267.8013,
"train_samples_per_second": 32.935,
"train_steps_per_second": 1.031
},
{
"epoch": 2.0,
"eval_accuracy": 0.8277777777777777,
"eval_f1": 0.8242008421588933,
"eval_loss": 0.8600034117698669,
"eval_precision": 0.8391484412975262,
"eval_recall": 0.8355373608358867,
"eval_runtime": 114.9843,
"eval_samples_per_second": 32.874,
"eval_steps_per_second": 1.035,
"step": 552
},
{
"epoch": 3.0,
"step": 828,
"train_eval_accuracy": 0.9051020408163265,
"train_eval_f1": 0.9035552222804545,
"train_eval_loss": 0.4452792704105377,
"train_eval_precision": 0.9086137959345946,
"train_eval_recall": 0.9041820807070131,
"train_loss": 0.4452792704105377,
"train_runtime": 267.5894,
"train_samples_per_second": 32.961,
"train_steps_per_second": 1.031
},
{
"epoch": 3.0,
"eval_accuracy": 0.8804232804232804,
"eval_f1": 0.8806958156074617,
"eval_loss": 0.5410670638084412,
"eval_precision": 0.8871882485699742,
"eval_recall": 0.8835379115156438,
"eval_runtime": 114.901,
"eval_samples_per_second": 32.898,
"eval_steps_per_second": 1.036,
"step": 828
},
{
"epoch": 4.0,
"step": 1104,
"train_eval_accuracy": 0.9252834467120181,
"train_eval_f1": 0.9244941105182266,
"train_eval_loss": 0.3236485421657562,
"train_eval_precision": 0.9275441941761358,
"train_eval_recall": 0.924775463750757,
"train_loss": 0.32364851236343384,
"train_runtime": 268.0076,
"train_samples_per_second": 32.91,
"train_steps_per_second": 1.03
},
{
"epoch": 4.0,
"eval_accuracy": 0.8992063492063492,
"eval_f1": 0.8994418745237454,
"eval_loss": 0.44155266880989075,
"eval_precision": 0.9029497697773854,
"eval_recall": 0.9012117948893439,
"eval_runtime": 114.5928,
"eval_samples_per_second": 32.986,
"eval_steps_per_second": 1.038,
"step": 1104
},
{
"epoch": 5.0,
"step": 1380,
"train_eval_accuracy": 0.9377551020408164,
"train_eval_f1": 0.9368640571091438,
"train_eval_loss": 0.2548055350780487,
"train_eval_precision": 0.9393905662960794,
"train_eval_recall": 0.9370066740159024,
"train_loss": 0.2548055350780487,
"train_runtime": 266.9807,
"train_samples_per_second": 33.036,
"train_steps_per_second": 1.034
},
{
"epoch": 5.0,
"eval_accuracy": 0.9047619047619048,
"eval_f1": 0.9049060279992467,
"eval_loss": 0.4057652950286865,
"eval_precision": 0.9074489392634286,
"eval_recall": 0.9071413991393202,
"eval_runtime": 114.474,
"eval_samples_per_second": 33.021,
"eval_steps_per_second": 1.04,
"step": 1380
},
{
"epoch": 6.0,
"step": 1656,
"train_eval_accuracy": 0.9470521541950113,
"train_eval_f1": 0.9465820465281242,
"train_eval_loss": 0.21100522577762604,
"train_eval_precision": 0.9475994271224741,
"train_eval_recall": 0.9466568764400392,
"train_loss": 0.21100524067878723,
"train_runtime": 267.4217,
"train_samples_per_second": 32.982,
"train_steps_per_second": 1.032
},
{
"epoch": 6.0,
"eval_accuracy": 0.9097883597883598,
"eval_f1": 0.9104147506557622,
"eval_loss": 0.38457947969436646,
"eval_precision": 0.9123207267374824,
"eval_recall": 0.9116754514446468,
"eval_runtime": 114.6488,
"eval_samples_per_second": 32.97,
"eval_steps_per_second": 1.038,
"step": 1656
},
{
"epoch": 7.0,
"step": 1932,
"train_eval_accuracy": 0.9549886621315192,
"train_eval_f1": 0.9546212565256129,
"train_eval_loss": 0.1847783774137497,
"train_eval_precision": 0.9559007170334742,
"train_eval_recall": 0.9545838862429018,
"train_loss": 0.1847783923149109,
"train_runtime": 266.8425,
"train_samples_per_second": 33.053,
"train_steps_per_second": 1.034
},
{
"epoch": 7.0,
"eval_accuracy": 0.9082010582010582,
"eval_f1": 0.908576534948754,
"eval_loss": 0.38885927200317383,
"eval_precision": 0.911182168262048,
"eval_recall": 0.9099497041696394,
"eval_runtime": 114.6761,
"eval_samples_per_second": 32.962,
"eval_steps_per_second": 1.038,
"step": 1932
},
{
"epoch": 8.0,
"step": 2208,
"train_eval_accuracy": 0.962358276643991,
"train_eval_f1": 0.9620445758202029,
"train_eval_loss": 0.15657010674476624,
"train_eval_precision": 0.9625235916902362,
"train_eval_recall": 0.9619882239348216,
"train_loss": 0.15657009184360504,
"train_runtime": 267.8869,
"train_samples_per_second": 32.924,
"train_steps_per_second": 1.03
},
{
"epoch": 8.0,
"eval_accuracy": 0.91005291005291,
"eval_f1": 0.9102033809946286,
"eval_loss": 0.38644978404045105,
"eval_precision": 0.9106144278267643,
"eval_recall": 0.9117775212470368,
"eval_runtime": 115.022,
"eval_samples_per_second": 32.863,
"eval_steps_per_second": 1.035,
"step": 2208
},
{
"epoch": 9.0,
"step": 2484,
"train_eval_accuracy": 0.9653061224489796,
"train_eval_f1": 0.9649036069253767,
"train_eval_loss": 0.13657130300998688,
"train_eval_precision": 0.966025675321776,
"train_eval_recall": 0.9649842543919382,
"train_loss": 0.13657130300998688,
"train_runtime": 267.8957,
"train_samples_per_second": 32.923,
"train_steps_per_second": 1.03
},
{
"epoch": 9.0,
"eval_accuracy": 0.9063492063492063,
"eval_f1": 0.9069930228990634,
"eval_loss": 0.4001522362232208,
"eval_precision": 0.909308614451706,
"eval_recall": 0.9081607205793748,
"eval_runtime": 114.6165,
"eval_samples_per_second": 32.98,
"eval_steps_per_second": 1.038,
"step": 2484
},
{
"epoch": 9.0,
"step": 2484,
"total_flos": 2.08932565653504e+16,
"train_loss": 0.7482529300806411,
"train_runtime": 11065.7342,
"train_samples_per_second": 15.941,
"train_steps_per_second": 0.499
},
{
"epoch": 9.0,
"eval_accuracy": 0.9063492063492063,
"eval_f1": 0.9069930228990634,
"eval_loss": 0.4001522362232208,
"eval_precision": 0.909308614451706,
"eval_recall": 0.9081607205793748,
"eval_runtime": 114.8255,
"eval_samples_per_second": 32.92,
"eval_steps_per_second": 1.036,
"step": 2484
},
{
"epoch": 9.0,
"step": 2484,
"train_en_eval_accuracy": 0.9653061224489796,
"train_en_eval_f1": 0.9649036069253767,
"train_en_eval_loss": 0.13657130300998688,
"train_en_eval_precision": 0.966025675321776,
"train_en_eval_recall": 0.9649842543919382,
"train_en_loss": 0.13657130300998688,
"train_en_runtime": 267.45,
"train_en_samples_per_second": 32.978,
"train_en_steps_per_second": 1.032
},
{
"epoch": 9.0,
"step": 2484,
"test_en_eval_accuracy": 0.9063492063492063,
"test_en_eval_f1": 0.9069930228990634,
"test_en_eval_loss": 0.40015220642089844,
"test_en_eval_precision": 0.909308614451706,
"test_en_eval_recall": 0.9081607205793748,
"test_en_loss": 0.4001522362232208,
"test_en_runtime": 114.5156,
"test_en_samples_per_second": 33.009,
"test_en_steps_per_second": 1.039
}
],
"logging_steps": 500,
"max_steps": 5520,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"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": 2.08932565653504e+16,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}