longformer-simple / checkpoint-810 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 10, checkpoint
c2b639e verified
raw
history blame
13.8 kB
{
"best_metric": 0.4725272059440613,
"best_model_checkpoint": "longformer-simple/checkpoint-324",
"epoch": 10.0,
"eval_steps": 500,
"global_step": 810,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_Claim": {
"f1-score": 0.46242900830056793,
"precision": 0.43257049448304047,
"recall": 0.496715157203191,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.5626477541371159,
"precision": 0.6522533495736906,
"recall": 0.49468822170900695,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8691687871077185,
"precision": 0.911293908403735,
"recall": 0.8307661126874747,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8615178272046495,
"precision": 0.8377046804810897,
"recall": 0.88672444205844,
"support": 13039.0
},
"eval_accuracy": 0.7823004022635849,
"eval_loss": 0.5488611459732056,
"eval_macro avg": {
"f1-score": 0.6889408441875129,
"precision": 0.708455608235389,
"recall": 0.6772234834145282,
"support": 29334.0
},
"eval_runtime": 5.8698,
"eval_samples_per_second": 13.629,
"eval_steps_per_second": 1.704,
"eval_weighted avg": {
"f1-score": 0.7840489998358311,
"precision": 0.7899101236188295,
"recall": 0.7823004022635849,
"support": 29334.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.5457327068079991,
"precision": 0.5107383923092657,
"recall": 0.5858751759737213,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.6722085773663329,
"precision": 0.5687340153452686,
"recall": 0.8217090069284064,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8773589977955805,
"precision": 0.9344863131370977,
"recall": 0.8268139440616133,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8678243902439025,
"precision": 0.8834419195931988,
"recall": 0.852749443975765,
"support": 13039.0
},
"eval_accuracy": 0.8029590236585532,
"eval_loss": 0.5075464248657227,
"eval_macro avg": {
"f1-score": 0.7407811680534537,
"precision": 0.7243501600962077,
"recall": 0.7717868927348766,
"support": 29334.0
},
"eval_runtime": 5.9473,
"eval_samples_per_second": 13.451,
"eval_steps_per_second": 1.681,
"eval_weighted avg": {
"f1-score": 0.8097969994222006,
"precision": 0.8232353684753937,
"recall": 0.8029590236585532,
"support": 29334.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.4978109708987896,
"precision": 0.5516552511415526,
"recall": 0.45354293758798686,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7357876283496118,
"precision": 0.8036105032822757,
"recall": 0.6785219399538106,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8891011355194736,
"precision": 0.9331699710403207,
"recall": 0.8490068909606809,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8770979581655561,
"precision": 0.8191560170394037,
"recall": 0.9438607255157604,
"support": 13039.0
},
"eval_accuracy": 0.821129065248517,
"eval_loss": 0.5219377279281616,
"eval_macro avg": {
"f1-score": 0.7499494232333578,
"precision": 0.7768979356258882,
"recall": 0.7312331235045597,
"support": 29334.0
},
"eval_runtime": 5.8364,
"eval_samples_per_second": 13.707,
"eval_steps_per_second": 1.713,
"eval_weighted avg": {
"f1-score": 0.815598992812927,
"precision": 0.8174973750724106,
"recall": 0.821129065248517,
"support": 29334.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.5811481218993623,
"precision": 0.5851569933396765,
"recall": 0.5771938057250118,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.761714412613813,
"precision": 0.7335329341317365,
"recall": 0.792147806004619,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8977645611156685,
"precision": 0.9085720215857203,
"recall": 0.8872111876773409,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8891009734682191,
"precision": 0.8851474612344178,
"recall": 0.8930899608865711,
"support": 13039.0
},
"eval_accuracy": 0.8377650507943001,
"eval_loss": 0.4725272059440613,
"eval_macro avg": {
"f1-score": 0.7824320172742658,
"precision": 0.7781023525728877,
"recall": 0.7874106900733857,
"support": 29334.0
},
"eval_runtime": 5.8763,
"eval_samples_per_second": 13.614,
"eval_steps_per_second": 1.702,
"eval_weighted avg": {
"f1-score": 0.8378705011585708,
"precision": 0.8382513248807655,
"recall": 0.8377650507943001,
"support": 29334.0
},
"step": 324
},
{
"epoch": 5.0,
"eval_Claim": {
"f1-score": 0.5790953831731895,
"precision": 0.5784176029962547,
"recall": 0.5797747536367902,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7450592885375493,
"precision": 0.8008497079129049,
"recall": 0.6965357967667436,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8897467572575665,
"precision": 0.9040794979079498,
"recall": 0.8758613700851237,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8861129867206842,
"precision": 0.8672442910639547,
"recall": 0.905820998542833,
"support": 13039.0
},
"eval_accuracy": 0.8329242517215518,
"eval_loss": 0.5538631081581116,
"eval_macro avg": {
"f1-score": 0.7750036039222473,
"precision": 0.787647774970266,
"recall": 0.7644982297578727,
"support": 29334.0
},
"eval_runtime": 5.8534,
"eval_samples_per_second": 13.667,
"eval_steps_per_second": 1.708,
"eval_weighted avg": {
"f1-score": 0.8323176558681596,
"precision": 0.8327711951367025,
"recall": 0.8329242517215518,
"support": 29334.0
},
"step": 405
},
{
"epoch": 6.0,
"eval_Claim": {
"f1-score": 0.5991994252283692,
"precision": 0.5325670498084292,
"recall": 0.6848897231346786,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7586672669968483,
"precision": 0.7400087834870444,
"recall": 0.7782909930715936,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8972759409440632,
"precision": 0.9212211784799317,
"recall": 0.8745439805431698,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8764605695726225,
"precision": 0.9062909567496723,
"recall": 0.8485313290896541,
"support": 13039.0
},
"eval_accuracy": 0.8283220835890094,
"eval_loss": 0.5790187120437622,
"eval_macro avg": {
"f1-score": 0.7829008006854759,
"precision": 0.7750219921312693,
"recall": 0.796564006459774,
"support": 29334.0
},
"eval_runtime": 5.8468,
"eval_samples_per_second": 13.683,
"eval_steps_per_second": 1.71,
"eval_weighted avg": {
"f1-score": 0.8344852708551485,
"precision": 0.8447418748493872,
"recall": 0.8283220835890094,
"support": 29334.0
},
"step": 486
},
{
"epoch": 6.17,
"grad_norm": 6.636661529541016,
"learning_rate": 1.3827160493827162e-05,
"loss": 0.3561,
"step": 500
},
{
"epoch": 7.0,
"eval_Claim": {
"f1-score": 0.5628712871287128,
"precision": 0.5955997904662127,
"recall": 0.5335523228531206,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7734199350487135,
"precision": 0.8422198041349293,
"recall": 0.7150115473441109,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.9010954940057875,
"precision": 0.9193378321383383,
"recall": 0.8835630320226996,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8842947894099071,
"precision": 0.848316189939411,
"recall": 0.9234603880665695,
"support": 13039.0
},
"eval_accuracy": 0.838003681734506,
"eval_loss": 0.7058164477348328,
"eval_macro avg": {
"f1-score": 0.7804203763982802,
"precision": 0.8013684041697228,
"recall": 0.7638968225716252,
"support": 29334.0
},
"eval_runtime": 5.8534,
"eval_samples_per_second": 13.667,
"eval_steps_per_second": 1.708,
"eval_weighted avg": {
"f1-score": 0.835063123988816,
"precision": 0.8350403187795808,
"recall": 0.838003681734506,
"support": 29334.0
},
"step": 567
},
{
"epoch": 8.0,
"eval_Claim": {
"f1-score": 0.5638466220328667,
"precision": 0.5858841386288894,
"recall": 0.5434068512435476,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7606318347509112,
"precision": 0.8025641025641026,
"recall": 0.7228637413394919,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.896161170692593,
"precision": 0.8864763037874281,
"recall": 0.9060599918929875,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8909187386294724,
"precision": 0.8807043836642937,
"recall": 0.9013728046629342,
"support": 13039.0
},
"eval_accuracy": 0.8377650507943001,
"eval_loss": 0.6875935196876526,
"eval_macro avg": {
"f1-score": 0.7778895915264609,
"precision": 0.7889072321611785,
"recall": 0.7684258472847404,
"support": 29334.0
},
"eval_runtime": 5.8627,
"eval_samples_per_second": 13.646,
"eval_steps_per_second": 1.706,
"eval_weighted avg": {
"f1-score": 0.8355454452418354,
"precision": 0.8340438435010799,
"recall": 0.8377650507943001,
"support": 29334.0
},
"step": 648
},
{
"epoch": 9.0,
"eval_Claim": {
"f1-score": 0.5931978798586572,
"precision": 0.5602836879432624,
"recall": 0.6302205537306429,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7739087788131438,
"precision": 0.8248823836905385,
"recall": 0.7288683602771363,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.8940459939094823,
"precision": 0.927653083460449,
"recall": 0.8627888123226591,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8843249773482331,
"precision": 0.8708454160160607,
"recall": 0.8982283917478334,
"support": 13039.0
},
"eval_accuracy": 0.8348673893775141,
"eval_loss": 0.6963120698928833,
"eval_macro avg": {
"f1-score": 0.786369407482379,
"precision": 0.7959161427775777,
"recall": 0.7800265295195679,
"support": 29334.0
},
"eval_runtime": 5.8671,
"eval_samples_per_second": 13.635,
"eval_steps_per_second": 1.704,
"eval_weighted avg": {
"f1-score": 0.8371473756606818,
"precision": 0.8414411074427396,
"recall": 0.8348673893775141,
"support": 29334.0
},
"step": 729
},
{
"epoch": 10.0,
"eval_Claim": {
"f1-score": 0.5676917638741604,
"precision": 0.5701775147928994,
"recall": 0.5652275926794932,
"support": 4262.0
},
"eval_MajorClaim": {
"f1-score": 0.7616811458088754,
"precision": 0.7745940783190067,
"recall": 0.7491916859122402,
"support": 2165.0
},
"eval_O": {
"f1-score": 0.9002143118498772,
"precision": 0.9296124365756234,
"recall": 0.8726185650587759,
"support": 9868.0
},
"eval_Premise": {
"f1-score": 0.8900003732596766,
"precision": 0.8669284467713787,
"recall": 0.9143339213129841,
"support": 13039.0
},
"eval_accuracy": 0.8373900593168337,
"eval_loss": 0.7715491056442261,
"eval_macro avg": {
"f1-score": 0.7798968986981474,
"precision": 0.785328119114727,
"recall": 0.7753429412408733,
"support": 29334.0
},
"eval_runtime": 5.8552,
"eval_samples_per_second": 13.663,
"eval_steps_per_second": 1.708,
"eval_weighted avg": {
"f1-score": 0.8371368267053725,
"precision": 0.8380850988337167,
"recall": 0.8373900593168337,
"support": 29334.0
},
"step": 810
}
],
"logging_steps": 500,
"max_steps": 1620,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 500,
"total_flos": 1437887327268000.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}