longformer-full_labels / checkpoint-533 /trainer_state.json
Theoreticallyhugo's picture
Training in progress, epoch 13, checkpoint
1663da5 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 13.0,
"eval_steps": 500,
"global_step": 533,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_B-Claim": {
"f1-score": 0.0,
"precision": 0.0,
"recall": 0.0,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.0,
"precision": 0.0,
"recall": 0.0,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.1227621483375959,
"precision": 0.8727272727272727,
"recall": 0.06602475928473177,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.35051546391752586,
"precision": 0.5111008325624422,
"recall": 0.26671494086410813,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.2476060191518468,
"precision": 0.4952120383036936,
"recall": 0.16507067943456452,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8531195999852935,
"precision": 0.7927029242962558,
"recall": 0.9235055321181247,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.8310506279263117,
"precision": 0.7493095557484416,
"recall": 0.9328094302554027,
"support": 10180.0
},
"eval_accuracy": 0.7473642462901147,
"eval_loss": 0.6794929504394531,
"eval_macro avg": {
"f1-score": 0.34357912275979624,
"precision": 0.48872180337687227,
"recall": 0.3363036202795617,
"support": 30257.0
},
"eval_runtime": 1.4284,
"eval_samples_per_second": 56.707,
"eval_steps_per_second": 7.701,
"eval_weighted avg": {
"f1-score": 0.7027223642713039,
"precision": 0.7080894203665903,
"recall": 0.7473642462901147,
"support": 30257.0
},
"step": 41
},
{
"epoch": 2.0,
"eval_B-Claim": {
"f1-score": 0.1337386018237082,
"precision": 0.6470588235294118,
"recall": 0.07457627118644068,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.0,
"precision": 0.0,
"recall": 0.0,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7234567901234568,
"precision": 0.6562150055991042,
"recall": 0.8060522696011004,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5708538220831472,
"precision": 0.5315296566077004,
"recall": 0.6164615013275404,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.6949243098842386,
"precision": 0.5701344243132671,
"recall": 0.8896488828089375,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.861198871056068,
"precision": 0.914638511095204,
"recall": 0.8136591578444639,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9030685830805192,
"precision": 0.9145764077767704,
"recall": 0.8918467583497053,
"support": 10180.0
},
"eval_accuracy": 0.8068876623591235,
"eval_loss": 0.511056661605835,
"eval_macro avg": {
"f1-score": 0.5553201397215911,
"precision": 0.6048789755602082,
"recall": 0.5846064058740269,
"support": 30257.0
},
"eval_runtime": 1.4332,
"eval_samples_per_second": 56.519,
"eval_steps_per_second": 7.675,
"eval_weighted avg": {
"f1-score": 0.8086360829977005,
"precision": 0.8236564850419075,
"recall": 0.8068876623591235,
"support": 30257.0
},
"step": 82
},
{
"epoch": 3.0,
"eval_B-Claim": {
"f1-score": 0.47176079734219273,
"precision": 0.46254071661237783,
"recall": 0.48135593220338985,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.07272727272727274,
"precision": 0.6666666666666666,
"recall": 0.038461538461538464,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7764857881136951,
"precision": 0.7320341047503045,
"recall": 0.8266850068775791,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.597796762589928,
"precision": 0.5594361455922575,
"recall": 0.6418054549843109,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.730778274159417,
"precision": 0.7543689320388349,
"recall": 0.7086183310533516,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8887291317197858,
"precision": 0.8918637274549098,
"recall": 0.8856164928759054,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9178492904257447,
"precision": 0.934092758340114,
"recall": 0.9021611001964637,
"support": 10180.0
},
"eval_accuracy": 0.835244736755131,
"eval_loss": 0.4231853783130646,
"eval_macro avg": {
"f1-score": 0.6365896167254336,
"precision": 0.7144290073507807,
"recall": 0.6406719795217912,
"support": 30257.0
},
"eval_runtime": 1.4362,
"eval_samples_per_second": 56.399,
"eval_steps_per_second": 7.659,
"eval_weighted avg": {
"f1-score": 0.836272553745959,
"precision": 0.841400720911604,
"recall": 0.835244736755131,
"support": 30257.0
},
"step": 123
},
{
"epoch": 4.0,
"eval_B-Claim": {
"f1-score": 0.5212947189097104,
"precision": 0.523972602739726,
"recall": 0.5186440677966102,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.6433566433566433,
"precision": 0.7076923076923077,
"recall": 0.5897435897435898,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7744020685197156,
"precision": 0.7304878048780488,
"recall": 0.8239339752407153,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5872821016449792,
"precision": 0.5975518361229079,
"recall": 0.5773594013999517,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.7907801418439716,
"precision": 0.7692971108236308,
"recall": 0.8134974920200638,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8883939418482223,
"precision": 0.9042123485285631,
"recall": 0.873119477831728,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9123517500723171,
"precision": 0.8958530581329294,
"recall": 0.9294695481335953,
"support": 10180.0
},
"eval_accuracy": 0.8411607231384473,
"eval_loss": 0.42953914403915405,
"eval_macro avg": {
"f1-score": 0.7311230523136514,
"precision": 0.7327238669883019,
"recall": 0.7322525074523221,
"support": 30257.0
},
"eval_runtime": 1.4336,
"eval_samples_per_second": 56.501,
"eval_steps_per_second": 7.673,
"eval_weighted avg": {
"f1-score": 0.8405678153025818,
"precision": 0.8407365647422267,
"recall": 0.8411607231384473,
"support": 30257.0
},
"step": 164
},
{
"epoch": 5.0,
"eval_B-Claim": {
"f1-score": 0.5539452495974235,
"precision": 0.5276073619631901,
"recall": 0.5830508474576271,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.6887417218543046,
"precision": 0.7123287671232876,
"recall": 0.6666666666666666,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.8093083387201035,
"precision": 0.7634146341463415,
"recall": 0.8610729023383769,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5725826193390453,
"precision": 0.5808294015396076,
"recall": 0.5645667390779628,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.7844767844767844,
"precision": 0.7949438202247191,
"recall": 0.774281805745554,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8950967891503419,
"precision": 0.8697251839615557,
"recall": 0.9219931545013134,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9191415785191211,
"precision": 0.9528679881906369,
"recall": 0.887721021611002,
"support": 10180.0
},
"eval_accuracy": 0.8447301450903923,
"eval_loss": 0.456537127494812,
"eval_macro avg": {
"f1-score": 0.7461847259510178,
"precision": 0.7431024510213342,
"recall": 0.7513361624855005,
"support": 30257.0
},
"eval_runtime": 1.4256,
"eval_samples_per_second": 56.819,
"eval_steps_per_second": 7.716,
"eval_weighted avg": {
"f1-score": 0.8445567746699831,
"precision": 0.8460194835144224,
"recall": 0.8447301450903923,
"support": 30257.0
},
"step": 205
},
{
"epoch": 6.0,
"eval_B-Claim": {
"f1-score": 0.5251798561151079,
"precision": 0.5593869731800766,
"recall": 0.49491525423728816,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7146974063400577,
"precision": 0.6492146596858639,
"recall": 0.7948717948717948,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7888040712468193,
"precision": 0.7337278106508875,
"recall": 0.8528198074277854,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5050387030816416,
"precision": 0.6394230769230769,
"recall": 0.41733043688148685,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8131487889273357,
"precision": 0.7733443027560675,
"recall": 0.8572731418148655,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8898820701569895,
"precision": 0.8424233805020266,
"recall": 0.9430072434927963,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9201684886169893,
"precision": 0.9398688793280066,
"recall": 0.9012770137524558,
"support": 10180.0
},
"eval_accuracy": 0.8434742373665598,
"eval_loss": 0.4916144609451294,
"eval_macro avg": {
"f1-score": 0.7367027692121344,
"precision": 0.7339127261465722,
"recall": 0.7516420989254959,
"support": 30257.0
},
"eval_runtime": 1.4434,
"eval_samples_per_second": 56.118,
"eval_steps_per_second": 7.621,
"eval_weighted avg": {
"f1-score": 0.8349273131912464,
"precision": 0.8360386273187937,
"recall": 0.8434742373665598,
"support": 30257.0
},
"step": 246
},
{
"epoch": 7.0,
"eval_B-Claim": {
"f1-score": 0.5845648604269295,
"precision": 0.5668789808917197,
"recall": 0.6033898305084746,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7261904761904762,
"precision": 0.6777777777777778,
"recall": 0.782051282051282,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.8031088082901554,
"precision": 0.758873929008568,
"recall": 0.8528198074277854,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5772882643372579,
"precision": 0.6070287539936102,
"recall": 0.5503258508327299,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8086917562724014,
"precision": 0.7948040510788199,
"recall": 0.8230734154126766,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8942490402140614,
"precision": 0.8718995765275257,
"recall": 0.9177744169386293,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9187421383647798,
"precision": 0.9417225373904075,
"recall": 0.8968565815324165,
"support": 10180.0
},
"eval_accuracy": 0.8482334666358198,
"eval_loss": 0.5145291686058044,
"eval_macro avg": {
"f1-score": 0.7589764777280089,
"precision": 0.745569372381204,
"recall": 0.7751844549577135,
"support": 30257.0
},
"eval_runtime": 1.4296,
"eval_samples_per_second": 56.659,
"eval_steps_per_second": 7.694,
"eval_weighted avg": {
"f1-score": 0.8468124537513947,
"precision": 0.8468453317066018,
"recall": 0.8482334666358198,
"support": 30257.0
},
"step": 287
},
{
"epoch": 8.0,
"eval_B-Claim": {
"f1-score": 0.5750394944707741,
"precision": 0.5384615384615384,
"recall": 0.6169491525423729,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7195121951219512,
"precision": 0.686046511627907,
"recall": 0.7564102564102564,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7879185817465528,
"precision": 0.7537688442211056,
"recall": 0.8253094910591472,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5927414588012767,
"precision": 0.5808619091751622,
"recall": 0.6051170649287956,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.7924528301886793,
"precision": 0.8207132388861749,
"recall": 0.7660738714090287,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8916528468767275,
"precision": 0.8846666144323435,
"recall": 0.8987502984955823,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9119720069982503,
"precision": 0.9284478371501272,
"recall": 0.8960707269155206,
"support": 10180.0
},
"eval_accuracy": 0.8427801830981261,
"eval_loss": 0.538555920124054,
"eval_macro avg": {
"f1-score": 0.7530413448863159,
"precision": 0.7418523562791942,
"recall": 0.7663829802515292,
"support": 30257.0
},
"eval_runtime": 1.4347,
"eval_samples_per_second": 56.458,
"eval_steps_per_second": 7.667,
"eval_weighted avg": {
"f1-score": 0.8439033621710555,
"precision": 0.8456178749149208,
"recall": 0.8427801830981261,
"support": 30257.0
},
"step": 328
},
{
"epoch": 9.0,
"eval_B-Claim": {
"f1-score": 0.5622895622895623,
"precision": 0.5585284280936454,
"recall": 0.5661016949152542,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7289156626506025,
"precision": 0.6875,
"recall": 0.7756410256410257,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7851186658114176,
"precision": 0.7355769230769231,
"recall": 0.8418156808803301,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.587080745341615,
"precision": 0.604811876119785,
"recall": 0.570359642770939,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8209806157354618,
"precision": 0.8211678832116789,
"recall": 0.8207934336525308,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8914104344376116,
"precision": 0.888625217528872,
"recall": 0.8942131656451484,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.917259306488793,
"precision": 0.91595650896268,
"recall": 0.918565815324165,
"support": 10180.0
},
"eval_accuracy": 0.8476716131804211,
"eval_loss": 0.5474137663841248,
"eval_macro avg": {
"f1-score": 0.7561507132507234,
"precision": 0.7445952624276548,
"recall": 0.7696414941184847,
"support": 30257.0
},
"eval_runtime": 1.4423,
"eval_samples_per_second": 56.16,
"eval_steps_per_second": 7.627,
"eval_weighted avg": {
"f1-score": 0.8467310896799954,
"precision": 0.8461371897103921,
"recall": 0.8476716131804211,
"support": 30257.0
},
"step": 369
},
{
"epoch": 10.0,
"eval_B-Claim": {
"f1-score": 0.5833333333333334,
"precision": 0.5531914893617021,
"recall": 0.6169491525423729,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7380281690140845,
"precision": 0.6582914572864321,
"recall": 0.8397435897435898,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.782495101241019,
"precision": 0.7450248756218906,
"recall": 0.8239339752407153,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5871350941488689,
"precision": 0.6161803713527851,
"recall": 0.5607048032826454,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8225702638918686,
"precision": 0.776742301458671,
"recall": 0.8741450068399452,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.888669171445521,
"precision": 0.8831158622858041,
"recall": 0.8942927644670858,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9133780094316207,
"precision": 0.9232313095835424,
"recall": 0.9037328094302554,
"support": 10180.0
},
"eval_accuracy": 0.8456555507816373,
"eval_loss": 0.5961998701095581,
"eval_macro avg": {
"f1-score": 0.7593727346437593,
"precision": 0.7365396667072609,
"recall": 0.7876431573638013,
"support": 30257.0
},
"eval_runtime": 1.4347,
"eval_samples_per_second": 56.458,
"eval_steps_per_second": 7.667,
"eval_weighted avg": {
"f1-score": 0.844598799846626,
"precision": 0.8446583764532496,
"recall": 0.8456555507816373,
"support": 30257.0
},
"step": 410
},
{
"epoch": 11.0,
"eval_B-Claim": {
"f1-score": 0.577922077922078,
"precision": 0.5545171339563862,
"recall": 0.6033898305084746,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7282913165266107,
"precision": 0.6467661691542289,
"recall": 0.8333333333333334,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7925283522348232,
"precision": 0.7694300518134715,
"recall": 0.8170563961485557,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.6015424164524421,
"precision": 0.6102831594634873,
"recall": 0.5930485155684286,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8221789040502492,
"precision": 0.783003300330033,
"recall": 0.8654810761513908,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8906343447230972,
"precision": 0.892235181338872,
"recall": 0.8890392422192152,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.917472118959108,
"precision": 0.9259629814907454,
"recall": 0.9091355599214146,
"support": 10180.0
},
"eval_accuracy": 0.8487622698879598,
"eval_loss": 0.5956980586051941,
"eval_macro avg": {
"f1-score": 0.7615099329812013,
"precision": 0.7403139967924607,
"recall": 0.7872119934072589,
"support": 30257.0
},
"eval_runtime": 1.4315,
"eval_samples_per_second": 56.583,
"eval_steps_per_second": 7.684,
"eval_weighted avg": {
"f1-score": 0.8488747259194414,
"precision": 0.8495500818518233,
"recall": 0.8487622698879598,
"support": 30257.0
},
"step": 451
},
{
"epoch": 12.0,
"eval_B-Claim": {
"f1-score": 0.5705024311183144,
"precision": 0.546583850931677,
"recall": 0.5966101694915255,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7277936962750716,
"precision": 0.6580310880829016,
"recall": 0.8141025641025641,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7944845699277742,
"precision": 0.7600502512562815,
"recall": 0.8321870701513068,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5785911255953873,
"precision": 0.6018252933507171,
"recall": 0.5570842384745354,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8082901554404146,
"precision": 0.7675276752767528,
"recall": 0.853625170998632,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8910157942416007,
"precision": 0.8818805551224076,
"recall": 0.9003422749343309,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9105163287725957,
"precision": 0.9259597806215722,
"recall": 0.8955795677799607,
"support": 10180.0
},
"eval_accuracy": 0.8433089863502661,
"eval_loss": 0.6362738013267517,
"eval_macro avg": {
"f1-score": 0.7544563001958797,
"precision": 0.73455121352033,
"recall": 0.7785044365618364,
"support": 30257.0
},
"eval_runtime": 1.4338,
"eval_samples_per_second": 56.492,
"eval_steps_per_second": 7.672,
"eval_weighted avg": {
"f1-score": 0.8425156264772145,
"precision": 0.8427252530453597,
"recall": 0.8433089863502661,
"support": 30257.0
},
"step": 492
},
{
"epoch": 12.2,
"learning_rate": 4.75609756097561e-06,
"loss": 0.3164,
"step": 500
},
{
"epoch": 13.0,
"eval_B-Claim": {
"f1-score": 0.5771144278606964,
"precision": 0.564935064935065,
"recall": 0.5898305084745763,
"support": 295.0
},
"eval_B-MajorClaim": {
"f1-score": 0.7350427350427349,
"precision": 0.6615384615384615,
"recall": 0.8269230769230769,
"support": 156.0
},
"eval_B-Premise": {
"f1-score": 0.7949901120632828,
"precision": 0.7632911392405063,
"recall": 0.8294360385144429,
"support": 727.0
},
"eval_I-Claim": {
"f1-score": 0.5925018377848567,
"precision": 0.6016421995521274,
"recall": 0.5836350470673425,
"support": 4143.0
},
"eval_I-MajorClaim": {
"f1-score": 0.8274802458296752,
"precision": 0.7977147693609818,
"recall": 0.8595531235750113,
"support": 2193.0
},
"eval_I-Premise": {
"f1-score": 0.8883437524728971,
"precision": 0.8831720556997876,
"recall": 0.8935763750696489,
"support": 12563.0
},
"eval_O": {
"f1-score": 0.9159642910578025,
"precision": 0.9303008813696687,
"recall": 0.9020628683693517,
"support": 10180.0
},
"eval_accuracy": 0.8466801070826585,
"eval_loss": 0.6392464637756348,
"eval_macro avg": {
"f1-score": 0.7616339145874208,
"precision": 0.743227795956657,
"recall": 0.7835738625704929,
"support": 30257.0
},
"eval_runtime": 1.4326,
"eval_samples_per_second": 56.541,
"eval_steps_per_second": 7.678,
"eval_weighted avg": {
"f1-score": 0.8466487613672259,
"precision": 0.8471598021098077,
"recall": 0.8466801070826585,
"support": 30257.0
},
"step": 533
}
],
"logging_steps": 500,
"max_steps": 656,
"num_input_tokens_seen": 0,
"num_train_epochs": 16,
"save_steps": 500,
"total_flos": 1863475346187000.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}