k4black's picture
Training in progress, step 14500
f184108
raw
history blame
15.5 kB
{
"best_metric": 0.7575966715812683,
"best_model_checkpoint": "/home2/s5431786/jb-internship/results/Salesforce-codet5-small-CodeXGLUE-CONCODE-w_special_tokens/checkpoint-12000",
"epoch": 4.64,
"global_step": 14500,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.16,
"learning_rate": 3.160588611644274e-05,
"loss": 2.7739,
"step": 500
},
{
"epoch": 0.16,
"eval_bleu": 0.11002405260920435,
"eval_exact_match": 0.136,
"eval_loss": 0.9681498408317566,
"eval_rouge1": 0.5324111683230927,
"eval_rouge2": 0.3516111966646144,
"eval_rougeL": 0.5130793573577244,
"eval_runtime": 23.6738,
"eval_samples_per_second": 84.482,
"eval_steps_per_second": 2.661,
"step": 500
},
{
"epoch": 0.32,
"learning_rate": 6.35956493921945e-05,
"loss": 0.8968,
"step": 1000
},
{
"epoch": 0.32,
"eval_bleu": 0.11661343796568427,
"eval_exact_match": 0.148,
"eval_loss": 0.8933660984039307,
"eval_rouge1": 0.5490713398258988,
"eval_rouge2": 0.3737481223472951,
"eval_rougeL": 0.5313359382235461,
"eval_runtime": 21.9288,
"eval_samples_per_second": 91.204,
"eval_steps_per_second": 2.873,
"step": 1000
},
{
"epoch": 0.48,
"learning_rate": 9.558541266794626e-05,
"loss": 0.8139,
"step": 1500
},
{
"epoch": 0.48,
"eval_bleu": 0.11428222214709638,
"eval_exact_match": 0.148,
"eval_loss": 0.8664323687553406,
"eval_rouge1": 0.5501019947483123,
"eval_rouge2": 0.37606920276228095,
"eval_rougeL": 0.5316473251371836,
"eval_runtime": 21.9364,
"eval_samples_per_second": 91.173,
"eval_steps_per_second": 2.872,
"step": 1500
},
{
"epoch": 0.64,
"learning_rate": 9.854818607471284e-05,
"loss": 0.7535,
"step": 2000
},
{
"epoch": 0.64,
"eval_bleu": 0.12118583002054595,
"eval_exact_match": 0.152,
"eval_loss": 0.8371005058288574,
"eval_rouge1": 0.5554072128917051,
"eval_rouge2": 0.37994480723664975,
"eval_rougeL": 0.5360569772635451,
"eval_runtime": 21.9828,
"eval_samples_per_second": 90.98,
"eval_steps_per_second": 2.866,
"step": 2000
},
{
"epoch": 0.8,
"learning_rate": 9.686394718226834e-05,
"loss": 0.7267,
"step": 2500
},
{
"epoch": 0.8,
"eval_bleu": 0.1221326088002123,
"eval_exact_match": 0.1555,
"eval_loss": 0.8262679576873779,
"eval_rouge1": 0.5599864774843732,
"eval_rouge2": 0.3892847835247521,
"eval_rougeL": 0.5434109433715582,
"eval_runtime": 22.0553,
"eval_samples_per_second": 90.681,
"eval_steps_per_second": 2.856,
"step": 2500
},
{
"epoch": 0.96,
"learning_rate": 9.517970828982384e-05,
"loss": 0.7044,
"step": 3000
},
{
"epoch": 0.96,
"eval_bleu": 0.12264834126055617,
"eval_exact_match": 0.161,
"eval_loss": 0.8118048310279846,
"eval_rouge1": 0.5584777889931278,
"eval_rouge2": 0.38793170420432566,
"eval_rougeL": 0.539366409663261,
"eval_runtime": 21.8691,
"eval_samples_per_second": 91.453,
"eval_steps_per_second": 2.881,
"step": 3000
},
{
"epoch": 1.12,
"learning_rate": 9.349546939737934e-05,
"loss": 0.6505,
"step": 3500
},
{
"epoch": 1.12,
"eval_bleu": 0.12561336503475642,
"eval_exact_match": 0.1575,
"eval_loss": 0.8004248142242432,
"eval_rouge1": 0.5604501817439145,
"eval_rouge2": 0.39030440934575783,
"eval_rougeL": 0.5425107709271633,
"eval_runtime": 21.9286,
"eval_samples_per_second": 91.205,
"eval_steps_per_second": 2.873,
"step": 3500
},
{
"epoch": 1.28,
"learning_rate": 9.181123050493482e-05,
"loss": 0.6301,
"step": 4000
},
{
"epoch": 1.28,
"eval_bleu": 0.12658818034969338,
"eval_exact_match": 0.1605,
"eval_loss": 0.7999853491783142,
"eval_rouge1": 0.5627685830925565,
"eval_rouge2": 0.391269622424489,
"eval_rougeL": 0.5451139508088971,
"eval_runtime": 22.0195,
"eval_samples_per_second": 90.828,
"eval_steps_per_second": 2.861,
"step": 4000
},
{
"epoch": 1.44,
"learning_rate": 9.012699161249032e-05,
"loss": 0.6199,
"step": 4500
},
{
"epoch": 1.44,
"eval_bleu": 0.12894396386527762,
"eval_exact_match": 0.163,
"eval_loss": 0.7873425483703613,
"eval_rouge1": 0.5653377621170572,
"eval_rouge2": 0.39748707891409585,
"eval_rougeL": 0.5469501947752343,
"eval_runtime": 21.8948,
"eval_samples_per_second": 91.346,
"eval_steps_per_second": 2.877,
"step": 4500
},
{
"epoch": 1.6,
"learning_rate": 8.844275272004582e-05,
"loss": 0.619,
"step": 5000
},
{
"epoch": 1.6,
"eval_bleu": 0.13044433567409516,
"eval_exact_match": 0.159,
"eval_loss": 0.7792118191719055,
"eval_rouge1": 0.5697500148728435,
"eval_rouge2": 0.3972375515095064,
"eval_rougeL": 0.5508657148123793,
"eval_runtime": 22.0783,
"eval_samples_per_second": 90.587,
"eval_steps_per_second": 2.853,
"step": 5000
},
{
"epoch": 1.76,
"learning_rate": 8.67585138276013e-05,
"loss": 0.604,
"step": 5500
},
{
"epoch": 1.76,
"eval_bleu": 0.12974945121652084,
"eval_exact_match": 0.164,
"eval_loss": 0.7772428393363953,
"eval_rouge1": 0.5689486656450822,
"eval_rouge2": 0.3977032244845097,
"eval_rougeL": 0.5513327295716377,
"eval_runtime": 21.7891,
"eval_samples_per_second": 91.789,
"eval_steps_per_second": 2.891,
"step": 5500
},
{
"epoch": 1.92,
"learning_rate": 8.50742749351568e-05,
"loss": 0.5911,
"step": 6000
},
{
"epoch": 1.92,
"eval_bleu": 0.13125298952345468,
"eval_exact_match": 0.162,
"eval_loss": 0.7715250253677368,
"eval_rouge1": 0.5661217549934918,
"eval_rouge2": 0.3993895966921206,
"eval_rougeL": 0.5492973852073642,
"eval_runtime": 21.9641,
"eval_samples_per_second": 91.058,
"eval_steps_per_second": 2.868,
"step": 6000
},
{
"epoch": 2.08,
"learning_rate": 8.33900360427123e-05,
"loss": 0.5661,
"step": 6500
},
{
"epoch": 2.08,
"eval_bleu": 0.13309212763109746,
"eval_exact_match": 0.157,
"eval_loss": 0.7687653303146362,
"eval_rouge1": 0.5699981387398212,
"eval_rouge2": 0.3973095013395418,
"eval_rougeL": 0.551455131759035,
"eval_runtime": 21.9964,
"eval_samples_per_second": 90.924,
"eval_steps_per_second": 2.864,
"step": 6500
},
{
"epoch": 2.24,
"learning_rate": 8.17057971502678e-05,
"loss": 0.5341,
"step": 7000
},
{
"epoch": 2.24,
"eval_bleu": 0.13384219047455637,
"eval_exact_match": 0.161,
"eval_loss": 0.7724990248680115,
"eval_rouge1": 0.5718059859082061,
"eval_rouge2": 0.40004342552261796,
"eval_rougeL": 0.5526613011742771,
"eval_runtime": 21.9605,
"eval_samples_per_second": 91.073,
"eval_steps_per_second": 2.869,
"step": 7000
},
{
"epoch": 2.4,
"learning_rate": 8.00215582578233e-05,
"loss": 0.5368,
"step": 7500
},
{
"epoch": 2.4,
"eval_bleu": 0.1330977205657701,
"eval_exact_match": 0.161,
"eval_loss": 0.7691414952278137,
"eval_rouge1": 0.5709709775537742,
"eval_rouge2": 0.40272100034764474,
"eval_rougeL": 0.5533607655778048,
"eval_runtime": 22.1482,
"eval_samples_per_second": 90.301,
"eval_steps_per_second": 2.844,
"step": 7500
},
{
"epoch": 2.56,
"learning_rate": 7.83373193653788e-05,
"loss": 0.5326,
"step": 8000
},
{
"epoch": 2.56,
"eval_bleu": 0.13382817289236076,
"eval_exact_match": 0.161,
"eval_loss": 0.7643744945526123,
"eval_rouge1": 0.5679296577675541,
"eval_rouge2": 0.40079835184768015,
"eval_rougeL": 0.5518721002250275,
"eval_runtime": 21.8173,
"eval_samples_per_second": 91.671,
"eval_steps_per_second": 2.888,
"step": 8000
},
{
"epoch": 2.72,
"learning_rate": 7.665308047293428e-05,
"loss": 0.5296,
"step": 8500
},
{
"epoch": 2.72,
"eval_bleu": 0.13382829744509098,
"eval_exact_match": 0.161,
"eval_loss": 0.7650749683380127,
"eval_rouge1": 0.5692675142813927,
"eval_rouge2": 0.4018288384287947,
"eval_rougeL": 0.5521131597944772,
"eval_runtime": 21.7709,
"eval_samples_per_second": 91.866,
"eval_steps_per_second": 2.894,
"step": 8500
},
{
"epoch": 2.88,
"learning_rate": 7.496884158048978e-05,
"loss": 0.5343,
"step": 9000
},
{
"epoch": 2.88,
"eval_bleu": 0.1355027033693307,
"eval_exact_match": 0.1635,
"eval_loss": 0.7636184096336365,
"eval_rouge1": 0.5679441721859034,
"eval_rouge2": 0.4050592356179844,
"eval_rougeL": 0.5514489474326643,
"eval_runtime": 21.8929,
"eval_samples_per_second": 91.354,
"eval_steps_per_second": 2.878,
"step": 9000
},
{
"epoch": 3.04,
"learning_rate": 7.328460268804527e-05,
"loss": 0.5109,
"step": 9500
},
{
"epoch": 3.04,
"eval_bleu": 0.13581137320800807,
"eval_exact_match": 0.165,
"eval_loss": 0.7763196229934692,
"eval_rouge1": 0.5710735563830984,
"eval_rouge2": 0.4076167525021497,
"eval_rougeL": 0.5546392278395981,
"eval_runtime": 21.9516,
"eval_samples_per_second": 91.11,
"eval_steps_per_second": 2.87,
"step": 9500
},
{
"epoch": 3.2,
"learning_rate": 7.160036379560077e-05,
"loss": 0.4847,
"step": 10000
},
{
"epoch": 3.2,
"eval_bleu": 0.13710307705006092,
"eval_exact_match": 0.1645,
"eval_loss": 0.7670872807502747,
"eval_rouge1": 0.5697703902827196,
"eval_rouge2": 0.404941700282087,
"eval_rougeL": 0.5520241279397228,
"eval_runtime": 21.8083,
"eval_samples_per_second": 91.708,
"eval_steps_per_second": 2.889,
"step": 10000
},
{
"epoch": 3.36,
"learning_rate": 6.991612490315627e-05,
"loss": 0.4791,
"step": 10500
},
{
"epoch": 3.36,
"eval_bleu": 0.1361107213351937,
"eval_exact_match": 0.162,
"eval_loss": 0.7662444114685059,
"eval_rouge1": 0.5729589320755841,
"eval_rouge2": 0.40808662503016657,
"eval_rougeL": 0.555974205030025,
"eval_runtime": 21.95,
"eval_samples_per_second": 91.116,
"eval_steps_per_second": 2.87,
"step": 10500
},
{
"epoch": 3.52,
"learning_rate": 6.823188601071177e-05,
"loss": 0.4765,
"step": 11000
},
{
"epoch": 3.52,
"eval_bleu": 0.13494157409269927,
"eval_exact_match": 0.163,
"eval_loss": 0.7643869519233704,
"eval_rouge1": 0.5739569530373543,
"eval_rouge2": 0.40836724097565735,
"eval_rougeL": 0.5571861839936938,
"eval_runtime": 22.0521,
"eval_samples_per_second": 90.694,
"eval_steps_per_second": 2.857,
"step": 11000
},
{
"epoch": 3.68,
"learning_rate": 6.654764711826725e-05,
"loss": 0.4791,
"step": 11500
},
{
"epoch": 3.68,
"eval_bleu": 0.1355371343991678,
"eval_exact_match": 0.1665,
"eval_loss": 0.7635162472724915,
"eval_rouge1": 0.5745793842335204,
"eval_rouge2": 0.41206862743691836,
"eval_rougeL": 0.5587644331139192,
"eval_runtime": 22.3535,
"eval_samples_per_second": 89.472,
"eval_steps_per_second": 2.818,
"step": 11500
},
{
"epoch": 3.84,
"learning_rate": 6.486340822582275e-05,
"loss": 0.4779,
"step": 12000
},
{
"epoch": 3.84,
"eval_bleu": 0.1349476481247351,
"eval_exact_match": 0.163,
"eval_loss": 0.7575966715812683,
"eval_rouge1": 0.5806974627802324,
"eval_rouge2": 0.4141474945527226,
"eval_rougeL": 0.5638708523820954,
"eval_runtime": 24.2543,
"eval_samples_per_second": 82.46,
"eval_steps_per_second": 2.597,
"step": 12000
},
{
"epoch": 4.0,
"learning_rate": 6.317916933337825e-05,
"loss": 0.4723,
"step": 12500
},
{
"epoch": 4.0,
"eval_bleu": 0.13724470187361304,
"eval_exact_match": 0.1605,
"eval_loss": 0.7616467475891113,
"eval_rouge1": 0.5756965634335589,
"eval_rouge2": 0.4100511997403594,
"eval_rougeL": 0.559460669435164,
"eval_runtime": 23.0958,
"eval_samples_per_second": 86.596,
"eval_steps_per_second": 2.728,
"step": 12500
},
{
"epoch": 4.16,
"learning_rate": 6.149493044093374e-05,
"loss": 0.4422,
"step": 13000
},
{
"epoch": 4.16,
"eval_bleu": 0.13537517858880133,
"eval_exact_match": 0.163,
"eval_loss": 0.7656739354133606,
"eval_rouge1": 0.5705462095820004,
"eval_rouge2": 0.40541664718394044,
"eval_rougeL": 0.5548493921872111,
"eval_runtime": 24.1852,
"eval_samples_per_second": 82.695,
"eval_steps_per_second": 2.605,
"step": 13000
},
{
"epoch": 4.32,
"learning_rate": 5.981069154848924e-05,
"loss": 0.443,
"step": 13500
},
{
"epoch": 4.32,
"eval_bleu": 0.13844237313177618,
"eval_exact_match": 0.163,
"eval_loss": 0.7602689266204834,
"eval_rouge1": 0.5739436350821147,
"eval_rouge2": 0.4082942402749472,
"eval_rougeL": 0.5570060194504324,
"eval_runtime": 21.9206,
"eval_samples_per_second": 91.238,
"eval_steps_per_second": 2.874,
"step": 13500
},
{
"epoch": 4.48,
"learning_rate": 5.812645265604474e-05,
"loss": 0.4311,
"step": 14000
},
{
"epoch": 4.48,
"eval_bleu": 0.13569351540543104,
"eval_exact_match": 0.165,
"eval_loss": 0.7679688930511475,
"eval_rouge1": 0.5740205440854335,
"eval_rouge2": 0.4062700743361892,
"eval_rougeL": 0.5566120809764695,
"eval_runtime": 21.9949,
"eval_samples_per_second": 90.93,
"eval_steps_per_second": 2.864,
"step": 14000
},
{
"epoch": 4.64,
"learning_rate": 5.6442213763600224e-05,
"loss": 0.4324,
"step": 14500
},
{
"epoch": 4.64,
"eval_bleu": 0.136753744419093,
"eval_exact_match": 0.1655,
"eval_loss": 0.7612979412078857,
"eval_rouge1": 0.5796564826859487,
"eval_rouge2": 0.4106247491015199,
"eval_rougeL": 0.5617575380728914,
"eval_runtime": 21.8303,
"eval_samples_per_second": 91.616,
"eval_steps_per_second": 2.886,
"step": 14500
}
],
"max_steps": 31250,
"num_train_epochs": 10,
"total_flos": 5.235255191430758e+16,
"trial_name": null,
"trial_params": null
}