longformer-simple / checkpoint-324 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 4, checkpoint
d7783dc verified
raw
history blame
5.72 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 4.0,
"eval_steps": 500,
"global_step": 324,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_Claim": {
"f1-score": 0.4888068880688806,
"precision": 0.6228840125391849,
"recall": 0.40222672064777326,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.7325278370054489,
"precision": 0.7604525332021643,
"recall": 0.706581352833638,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.8634848138844484,
"precision": 0.827613377692173,
"recall": 0.902606702950444,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8792819197655248,
"precision": 0.8543458914268763,
"recall": 0.9057173407132524,
"support": 15899.0
},
"eval_accuracy": 0.8174925373134329,
"eval_loss": 0.49806877970695496,
"eval_macro avg": {
"f1-score": 0.7410253646810757,
"precision": 0.7663239537150996,
"recall": 0.7292830292862769,
"support": 33500.0
},
"eval_runtime": 4.9194,
"eval_samples_per_second": 16.465,
"eval_steps_per_second": 2.236,
"eval_weighted avg": {
"f1-score": 0.8071778099281224,
"precision": 0.80572410140167,
"recall": 0.8174925373134329,
"support": 33500.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.5993816254416962,
"precision": 0.6593780369290574,
"recall": 0.5493927125506073,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8017817371937639,
"precision": 0.7197382769901853,
"recall": 0.9049360146252285,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9058997050147494,
"precision": 0.9337184554575859,
"recall": 0.8796906330564308,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8854737486606459,
"precision": 0.862579028987236,
"recall": 0.9096169570413234,
"support": 15899.0
},
"eval_accuracy": 0.8468358208955223,
"eval_loss": 0.4562840759754181,
"eval_macro avg": {
"f1-score": 0.7981342040777137,
"precision": 0.7938534495910161,
"recall": 0.8109090793183975,
"support": 33500.0
},
"eval_runtime": 4.904,
"eval_samples_per_second": 16.517,
"eval_steps_per_second": 2.243,
"eval_weighted avg": {
"f1-score": 0.8442053257085675,
"precision": 0.8455251408465503,
"recall": 0.8468358208955223,
"support": 33500.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.6235462150581514,
"precision": 0.6285479226655697,
"recall": 0.6186234817813765,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8338848444738584,
"precision": 0.8059701492537313,
"recall": 0.8638025594149908,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9066250826953974,
"precision": 0.8974646833193002,
"recall": 0.9159744103886184,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8827730692315017,
"precision": 0.8911176621379133,
"recall": 0.8745833071262343,
"support": 15899.0
},
"eval_accuracy": 0.8490746268656716,
"eval_loss": 0.43820467591285706,
"eval_macro avg": {
"f1-score": 0.8117073028647273,
"precision": 0.8057751043441286,
"recall": 0.818245939677805,
"support": 33500.0
},
"eval_runtime": 4.9179,
"eval_samples_per_second": 16.47,
"eval_steps_per_second": 2.237,
"eval_weighted avg": {
"f1-score": 0.8488105033097496,
"precision": 0.8488213959184715,
"recall": 0.8490746268656716,
"support": 33500.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.6415287144900879,
"precision": 0.6477507222451506,
"recall": 0.6354251012145749,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8005126014523707,
"precision": 0.7514033680834001,
"recall": 0.856489945155393,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9065169636319258,
"precision": 0.9273871354374751,
"recall": 0.8865654540246348,
"support": 10473.0
},
"eval_Premise": {
"f1-score": 0.8932505382719131,
"precision": 0.8863636363636364,
"recall": 0.9002452984464432,
"support": 15899.0
},
"eval_accuracy": 0.8540597014925373,
"eval_loss": 0.4749625325202942,
"eval_macro avg": {
"f1-score": 0.8104522044615743,
"precision": 0.8032262155324155,
"recall": 0.8196814497102615,
"support": 33500.0
},
"eval_runtime": 4.8972,
"eval_samples_per_second": 16.54,
"eval_steps_per_second": 2.246,
"eval_weighted avg": {
"f1-score": 0.8542213698406307,
"precision": 0.8551874645146165,
"recall": 0.8540597014925373,
"support": 33500.0
},
"step": 324
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 573368735469600.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}