longformer-sep_tok / checkpoint-486 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 6, checkpoint
cacb729 verified
raw
history blame
8.35 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 6.0,
"eval_steps": 500,
"global_step": 486,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_Claim": {
"f1-score": 0.5385059461955894,
"precision": 0.6267132491265789,
"recall": 0.4720647773279352,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.7814784727863526,
"precision": 0.9588039867109635,
"recall": 0.6595063985374772,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9938608715302625,
"precision": 0.9996969237763298,
"recall": 0.9880925634688834,
"support": 13353.0
},
"eval_Premise": {
"f1-score": 0.8968837690149166,
"precision": 0.8455112497215416,
"recall": 0.954902824076986,
"support": 15899.0
},
"eval_accuracy": 0.8837548103353491,
"eval_loss": 0.2860471308231354,
"eval_macro avg": {
"f1-score": 0.8026822648817803,
"precision": 0.8576813523338533,
"recall": 0.7686416408528204,
"support": 36380.0
},
"eval_runtime": 4.9572,
"eval_samples_per_second": 16.34,
"eval_steps_per_second": 2.219,
"eval_weighted avg": {
"f1-score": 0.8768739289107891,
"precision": 0.8792073379911217,
"recall": 0.8837548103353491,
"support": 36380.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.6469459868753155,
"precision": 0.6453172205438067,
"recall": 0.648582995951417,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8435970110796187,
"precision": 0.9669226225634967,
"recall": 0.7481718464351006,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.998877245508982,
"precision": 0.9983541557567143,
"recall": 0.9994008836965476,
"support": 13353.0
},
"eval_Premise": {
"f1-score": 0.9099026477336144,
"precision": 0.8972179761540813,
"recall": 0.922951129001824,
"support": 15899.0
},
"eval_accuracy": 0.903243540406817,
"eval_loss": 0.24885737895965576,
"eval_macro avg": {
"f1-score": 0.8498307227993828,
"precision": 0.8769529937545246,
"recall": 0.8297767137712222,
"support": 36380.0
},
"eval_runtime": 5.0097,
"eval_samples_per_second": 16.169,
"eval_steps_per_second": 2.196,
"eval_weighted avg": {
"f1-score": 0.9028656814459712,
"precision": 0.9043261520711511,
"recall": 0.903243540406817,
"support": 36380.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.7040630102775911,
"precision": 0.6698958142935478,
"recall": 0.7419028340080972,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.8888411094388303,
"precision": 0.8392204628501827,
"recall": 0.9446983546617916,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.998989104796136,
"precision": 0.9988769092542678,
"recall": 0.9991013255448213,
"support": 13353.0
},
"eval_Premise": {
"f1-score": 0.914582593823615,
"precision": 0.9390987408880053,
"recall": 0.8913139191144097,
"support": 15899.0
},
"eval_accuracy": 0.9137987905442551,
"eval_loss": 0.24298855662345886,
"eval_macro avg": {
"f1-score": 0.8766189545840432,
"precision": 0.8617729818215009,
"recall": 0.8942541083322799,
"support": 36380.0
},
"eval_runtime": 4.9876,
"eval_samples_per_second": 16.24,
"eval_steps_per_second": 2.205,
"eval_weighted avg": {
"f1-score": 0.9154289662937856,
"precision": 0.9184781183611038,
"recall": 0.9137987905442551,
"support": 36380.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.6981611893583725,
"precision": 0.6754352763058289,
"recall": 0.7224696356275304,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.901231802911534,
"precision": 0.8836187966622749,
"recall": 0.9195612431444241,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9989130842172332,
"precision": 0.9998499399759904,
"recall": 0.9979779824758481,
"support": 13353.0
},
"eval_Premise": {
"f1-score": 0.9146224912392481,
"precision": 0.9266670970240785,
"recall": 0.9028869740235235,
"support": 15899.0
},
"eval_accuracy": 0.9142935678944475,
"eval_loss": 0.23980534076690674,
"eval_macro avg": {
"f1-score": 0.8782321419315969,
"precision": 0.8713927774920431,
"recall": 0.8857239588178316,
"support": 36380.0
},
"eval_runtime": 4.9726,
"eval_samples_per_second": 16.289,
"eval_steps_per_second": 2.212,
"eval_weighted avg": {
"f1-score": 0.9153623106642749,
"precision": 0.9168247558035478,
"recall": 0.9142935678944475,
"support": 36380.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_Claim": {
"f1-score": 0.6650318783717509,
"precision": 0.6450999048525214,
"recall": 0.6862348178137652,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.9172031076581575,
"precision": 0.8916702632714717,
"recall": 0.9442413162705667,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9982383147794145,
"precision": 0.9992495872730002,
"recall": 0.9972290870965326,
"support": 13353.0
},
"eval_Premise": {
"f1-score": 0.8998438545616774,
"precision": 0.9119622787753521,
"recall": 0.8880432731618341,
"support": 15899.0
},
"eval_accuracy": 0.9040956569543706,
"eval_loss": 0.2932502031326294,
"eval_macro avg": {
"f1-score": 0.87007928884275,
"precision": 0.8619955085430864,
"recall": 0.8789371235856747,
"support": 36380.0
},
"eval_runtime": 4.9595,
"eval_samples_per_second": 16.332,
"eval_steps_per_second": 2.218,
"eval_weighted avg": {
"f1-score": 0.9051179642561333,
"precision": 0.9065430476942037,
"recall": 0.9040956569543706,
"support": 36380.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_Claim": {
"f1-score": 0.6622698072805139,
"precision": 0.5740163325909429,
"recall": 0.7825910931174089,
"support": 4940.0
},
"eval_MajorClaim": {
"f1-score": 0.9223704351629127,
"precision": 0.8842767295597485,
"recall": 0.9638939670932358,
"support": 2188.0
},
"eval_O": {
"f1-score": 0.9982004948639124,
"precision": 0.9993994444861497,
"recall": 0.9970044184827379,
"support": 13353.0
},
"eval_Premise": {
"f1-score": 0.8780749379985253,
"precision": 0.9398091685199799,
"recall": 0.8239511918988616,
"support": 15899.0
},
"eval_accuracy": 0.8902693787795491,
"eval_loss": 0.3604726195335388,
"eval_macro avg": {
"f1-score": 0.865228918826466,
"precision": 0.8493754187892053,
"recall": 0.8918601676480611,
"support": 36380.0
},
"eval_runtime": 4.991,
"eval_samples_per_second": 16.229,
"eval_steps_per_second": 2.204,
"eval_weighted avg": {
"f1-score": 0.8955262233990811,
"precision": 0.9086708334194257,
"recall": 0.8902693787795491,
"support": 36380.0
},
"step": 486
}
],
"logging_steps": 500,
"max_steps": 4050,
"num_input_tokens_seen": 0,
"num_train_epochs": 50,
"save_steps": 500,
"total_flos": 860053103204400.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}