longformer-sep_tok / checkpoint-324 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 4, checkpoint
4f7d95d verified
raw
history blame
5.71 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": 0.46243291592128805,
"precision": 0.6588785046728972,
"recall": 0.3562241616903996,
"support": 4354.0
},
"eval_MajorClaim": {
"f1-score": 0.8342621259029929,
"precision": 0.8096955128205128,
"recall": 0.8603661132396765,
"support": 2349.0
},
"eval_O": {
"f1-score": 0.9993622448979592,
"precision": 0.9987253027405991,
"recall": 1.0,
"support": 12536.0
},
"eval_Premise": {
"f1-score": 0.8941052999825083,
"precision": 0.8401157057392676,
"recall": 0.955510692388216,
"support": 13374.0
},
"eval_accuracy": 0.8857510808573268,
"eval_loss": 0.2754688858985901,
"eval_macro avg": {
"f1-score": 0.7975406466761871,
"precision": 0.8268537564933192,
"recall": 0.793025241829573,
"support": 32613.0
},
"eval_runtime": 4.8781,
"eval_samples_per_second": 16.605,
"eval_steps_per_second": 2.255,
"eval_weighted avg": {
"f1-score": 0.8726239240080429,
"precision": 0.8746959682542021,
"recall": 0.8857510808573268,
"support": 32613.0
},
"step": 81
},
{
"epoch": 2.0,
"eval_Claim": {
"f1-score": 0.6446312571646924,
"precision": 0.5518482172064115,
"recall": 0.77491961414791,
"support": 4354.0
},
"eval_MajorClaim": {
"f1-score": 0.8912341158733577,
"precision": 0.9019180470793374,
"recall": 0.8808003405704555,
"support": 2349.0
},
"eval_O": {
"f1-score": 0.9998005504806733,
"precision": 0.9999202106439001,
"recall": 0.9996809189534142,
"support": 12536.0
},
"eval_Premise": {
"f1-score": 0.8690409646250897,
"precision": 0.9324023303632625,
"recall": 0.8137430835950351,
"support": 13374.0
},
"eval_accuracy": 0.8848618649004998,
"eval_loss": 0.25178349018096924,
"eval_macro avg": {
"f1-score": 0.8511767220359533,
"precision": 0.8465222013232279,
"recall": 0.8672859893167036,
"support": 32613.0
},
"eval_runtime": 4.907,
"eval_samples_per_second": 16.507,
"eval_steps_per_second": 2.242,
"eval_weighted avg": {
"f1-score": 0.8909418634778542,
"precision": 0.9053537287957648,
"recall": 0.8848618649004998,
"support": 32613.0
},
"step": 162
},
{
"epoch": 3.0,
"eval_Claim": {
"f1-score": 0.6598764019226366,
"precision": 0.6576186131386861,
"recall": 0.6621497473587505,
"support": 4354.0
},
"eval_MajorClaim": {
"f1-score": 0.9062119366626066,
"precision": 0.8661233993015134,
"recall": 0.9501915708812261,
"support": 2349.0
},
"eval_O": {
"f1-score": 0.9998404722022812,
"precision": 0.9997607273887382,
"recall": 0.9999202297383536,
"support": 12536.0
},
"eval_Premise": {
"f1-score": 0.9007852612503775,
"precision": 0.9097148086014946,
"recall": 0.8920293106026619,
"support": 13374.0
},
"eval_accuracy": 0.9070002759635728,
"eval_loss": 0.23778581619262695,
"eval_macro avg": {
"f1-score": 0.8666785180094754,
"precision": 0.8583043871076081,
"recall": 0.876072714645248,
"support": 32613.0
},
"eval_runtime": 4.9168,
"eval_samples_per_second": 16.474,
"eval_steps_per_second": 2.237,
"eval_weighted avg": {
"f1-score": 0.907089072967282,
"precision": 0.9075314026721463,
"recall": 0.9070002759635728,
"support": 32613.0
},
"step": 243
},
{
"epoch": 4.0,
"eval_Claim": {
"f1-score": 0.6604639922667954,
"precision": 0.696838347781744,
"recall": 0.6276986678915939,
"support": 4354.0
},
"eval_MajorClaim": {
"f1-score": 0.9140214420853479,
"precision": 0.9028239202657807,
"recall": 0.9255002128565347,
"support": 2349.0
},
"eval_O": {
"f1-score": 0.9994417862838917,
"precision": 0.9991230867346939,
"recall": 0.9997606892150607,
"support": 12536.0
},
"eval_Premise": {
"f1-score": 0.9084203149780549,
"precision": 0.8963534463934785,
"recall": 0.9208165096455809,
"support": 13374.0
},
"eval_accuracy": 0.912366234323736,
"eval_loss": 0.2598011791706085,
"eval_macro avg": {
"f1-score": 0.8705868839035225,
"precision": 0.8737847002939243,
"recall": 0.8684440199021926,
"support": 32613.0
},
"eval_runtime": 4.8872,
"eval_samples_per_second": 16.574,
"eval_steps_per_second": 2.251,
"eval_weighted avg": {
"f1-score": 0.9107077581074873,
"precision": 0.9096864919608297,
"recall": 0.912366234323736,
"support": 32613.0
},
"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
}