Md Mushfiqur Rahman
Upload with huggingface_hub
887d430
{
"best_metric": 0.8020833333333334,
"best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/canine/canine-base-finetuned-masakhaner-hau/checkpoint-7000",
"epoch": 133.33333333333334,
"global_step": 8000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 3.33,
"eval_accuracy_score": 0.9178235002478929,
"eval_f1": 0.44129032258064516,
"eval_loss": 0.2975754737854004,
"eval_precision": 0.38212290502793295,
"eval_recall": 0.5221374045801527,
"eval_runtime": 2.1678,
"eval_samples_per_second": 127.318,
"eval_steps_per_second": 16.145,
"step": 200
},
{
"epoch": 6.67,
"eval_accuracy_score": 0.9432325235498265,
"eval_f1": 0.625340599455041,
"eval_loss": 0.21037007868289948,
"eval_precision": 0.5645756457564576,
"eval_recall": 0.7007633587786259,
"eval_runtime": 2.1641,
"eval_samples_per_second": 127.538,
"eval_steps_per_second": 16.173,
"step": 400
},
{
"epoch": 8.33,
"learning_rate": 4.865771812080537e-05,
"loss": 0.3565,
"step": 500
},
{
"epoch": 10.0,
"eval_accuracy_score": 0.9494298463063956,
"eval_f1": 0.6824324324324323,
"eval_loss": 0.19267599284648895,
"eval_precision": 0.6121212121212121,
"eval_recall": 0.7709923664122137,
"eval_runtime": 2.1632,
"eval_samples_per_second": 127.587,
"eval_steps_per_second": 16.18,
"step": 600
},
{
"epoch": 13.33,
"eval_accuracy_score": 0.9559990084283589,
"eval_f1": 0.7209631728045326,
"eval_loss": 0.19005168974399567,
"eval_precision": 0.6723910171730515,
"eval_recall": 0.7770992366412214,
"eval_runtime": 2.1631,
"eval_samples_per_second": 127.597,
"eval_steps_per_second": 16.181,
"step": 800
},
{
"epoch": 16.67,
"learning_rate": 4.697986577181208e-05,
"loss": 0.0821,
"step": 1000
},
{
"epoch": 16.67,
"eval_accuracy_score": 0.9621963311849281,
"eval_f1": 0.7534148094895758,
"eval_loss": 0.17941339313983917,
"eval_precision": 0.7119565217391305,
"eval_recall": 0.8,
"eval_runtime": 2.1519,
"eval_samples_per_second": 128.257,
"eval_steps_per_second": 16.265,
"step": 1000
},
{
"epoch": 20.0,
"eval_accuracy_score": 0.9615765989092712,
"eval_f1": 0.7540751240255138,
"eval_loss": 0.18701964616775513,
"eval_precision": 0.7037037037037037,
"eval_recall": 0.8122137404580153,
"eval_runtime": 2.1592,
"eval_samples_per_second": 127.827,
"eval_steps_per_second": 16.21,
"step": 1200
},
{
"epoch": 23.33,
"eval_accuracy_score": 0.9639315815567675,
"eval_f1": 0.768892149669846,
"eval_loss": 0.20827054977416992,
"eval_precision": 0.7401129943502824,
"eval_recall": 0.8,
"eval_runtime": 2.1601,
"eval_samples_per_second": 127.772,
"eval_steps_per_second": 16.203,
"step": 1400
},
{
"epoch": 25.0,
"learning_rate": 4.530201342281879e-05,
"loss": 0.0161,
"step": 1500
},
{
"epoch": 26.67,
"eval_accuracy_score": 0.9636836886465047,
"eval_f1": 0.7759882869692533,
"eval_loss": 0.21991053223609924,
"eval_precision": 0.7454289732770746,
"eval_recall": 0.8091603053435115,
"eval_runtime": 2.1585,
"eval_samples_per_second": 127.866,
"eval_steps_per_second": 16.215,
"step": 1600
},
{
"epoch": 30.0,
"eval_accuracy_score": 0.9654189390183441,
"eval_f1": 0.7798833819241983,
"eval_loss": 0.20790381729602814,
"eval_precision": 0.7461645746164575,
"eval_recall": 0.816793893129771,
"eval_runtime": 2.1597,
"eval_samples_per_second": 127.793,
"eval_steps_per_second": 16.206,
"step": 1800
},
{
"epoch": 33.33,
"learning_rate": 4.36241610738255e-05,
"loss": 0.0063,
"step": 2000
},
{
"epoch": 33.33,
"eval_accuracy_score": 0.964427367377293,
"eval_f1": 0.7772241992882563,
"eval_loss": 0.22246131300926208,
"eval_precision": 0.728,
"eval_recall": 0.833587786259542,
"eval_runtime": 2.1499,
"eval_samples_per_second": 128.377,
"eval_steps_per_second": 16.28,
"step": 2000
},
{
"epoch": 36.67,
"eval_accuracy_score": 0.9641794744670302,
"eval_f1": 0.7784342688330871,
"eval_loss": 0.2224682867527008,
"eval_precision": 0.753934191702432,
"eval_recall": 0.8045801526717558,
"eval_runtime": 2.1602,
"eval_samples_per_second": 127.764,
"eval_steps_per_second": 16.202,
"step": 2200
},
{
"epoch": 40.0,
"eval_accuracy_score": 0.9669062964799207,
"eval_f1": 0.7894736842105263,
"eval_loss": 0.21754750609397888,
"eval_precision": 0.7573632538569425,
"eval_recall": 0.8244274809160306,
"eval_runtime": 2.1638,
"eval_samples_per_second": 127.555,
"eval_steps_per_second": 16.175,
"step": 2400
},
{
"epoch": 41.67,
"learning_rate": 4.194630872483222e-05,
"loss": 0.0036,
"step": 2500
},
{
"epoch": 43.33,
"eval_accuracy_score": 0.9652949925632127,
"eval_f1": 0.7895500725689405,
"eval_loss": 0.23659802973270416,
"eval_precision": 0.7524204702627939,
"eval_recall": 0.8305343511450382,
"eval_runtime": 2.1523,
"eval_samples_per_second": 128.234,
"eval_steps_per_second": 16.262,
"step": 2600
},
{
"epoch": 46.67,
"eval_accuracy_score": 0.9669062964799207,
"eval_f1": 0.7883636363636363,
"eval_loss": 0.2335628867149353,
"eval_precision": 0.7527777777777778,
"eval_recall": 0.8274809160305343,
"eval_runtime": 2.1466,
"eval_samples_per_second": 128.574,
"eval_steps_per_second": 16.305,
"step": 2800
},
{
"epoch": 50.0,
"learning_rate": 4.026845637583892e-05,
"loss": 0.0028,
"step": 3000
},
{
"epoch": 50.0,
"eval_accuracy_score": 0.9654189390183441,
"eval_f1": 0.7888970051132214,
"eval_loss": 0.24077844619750977,
"eval_precision": 0.7563025210084033,
"eval_recall": 0.8244274809160306,
"eval_runtime": 2.1522,
"eval_samples_per_second": 128.243,
"eval_steps_per_second": 16.263,
"step": 3000
},
{
"epoch": 53.33,
"eval_accuracy_score": 0.9655428854734754,
"eval_f1": 0.7809948032665183,
"eval_loss": 0.2439465969800949,
"eval_precision": 0.7601156069364162,
"eval_recall": 0.8030534351145038,
"eval_runtime": 2.1578,
"eval_samples_per_second": 127.91,
"eval_steps_per_second": 16.22,
"step": 3200
},
{
"epoch": 56.67,
"eval_accuracy_score": 0.9659147248388696,
"eval_f1": 0.7880597014925373,
"eval_loss": 0.2444428950548172,
"eval_precision": 0.7708029197080292,
"eval_recall": 0.8061068702290076,
"eval_runtime": 2.1536,
"eval_samples_per_second": 128.155,
"eval_steps_per_second": 16.252,
"step": 3400
},
{
"epoch": 58.33,
"learning_rate": 3.859060402684564e-05,
"loss": 0.0019,
"step": 3500
},
{
"epoch": 60.0,
"eval_accuracy_score": 0.9656668319286068,
"eval_f1": 0.7766423357664234,
"eval_loss": 0.24506314098834991,
"eval_precision": 0.7440559440559441,
"eval_recall": 0.8122137404580153,
"eval_runtime": 2.1569,
"eval_samples_per_second": 127.959,
"eval_steps_per_second": 16.227,
"step": 3600
},
{
"epoch": 63.33,
"eval_accuracy_score": 0.9675260287555776,
"eval_f1": 0.8005908419497784,
"eval_loss": 0.2420436143875122,
"eval_precision": 0.7753934191702432,
"eval_recall": 0.8274809160305343,
"eval_runtime": 2.151,
"eval_samples_per_second": 128.313,
"eval_steps_per_second": 16.272,
"step": 3800
},
{
"epoch": 66.67,
"learning_rate": 3.6912751677852356e-05,
"loss": 0.0024,
"step": 4000
},
{
"epoch": 66.67,
"eval_accuracy_score": 0.9659147248388696,
"eval_f1": 0.7849779086892489,
"eval_loss": 0.251669317483902,
"eval_precision": 0.7581792318634424,
"eval_recall": 0.8137404580152672,
"eval_runtime": 2.1475,
"eval_samples_per_second": 128.521,
"eval_steps_per_second": 16.298,
"step": 4000
},
{
"epoch": 70.0,
"eval_accuracy_score": 0.9652949925632127,
"eval_f1": 0.7835051546391752,
"eval_loss": 0.25008824467658997,
"eval_precision": 0.7567567567567568,
"eval_recall": 0.8122137404580153,
"eval_runtime": 2.1537,
"eval_samples_per_second": 128.15,
"eval_steps_per_second": 16.251,
"step": 4200
},
{
"epoch": 73.33,
"eval_accuracy_score": 0.9656668319286068,
"eval_f1": 0.7784256559766763,
"eval_loss": 0.24835215508937836,
"eval_precision": 0.7447698744769874,
"eval_recall": 0.815267175572519,
"eval_runtime": 2.15,
"eval_samples_per_second": 128.372,
"eval_steps_per_second": 16.279,
"step": 4400
},
{
"epoch": 75.0,
"learning_rate": 3.523489932885906e-05,
"loss": 0.002,
"step": 4500
},
{
"epoch": 76.67,
"eval_accuracy_score": 0.9655428854734754,
"eval_f1": 0.777859778597786,
"eval_loss": 0.26140472292900085,
"eval_precision": 0.7528571428571429,
"eval_recall": 0.8045801526717558,
"eval_runtime": 2.1498,
"eval_samples_per_second": 128.383,
"eval_steps_per_second": 16.28,
"step": 4600
},
{
"epoch": 80.0,
"eval_accuracy_score": 0.9669062964799207,
"eval_f1": 0.7905951506245408,
"eval_loss": 0.24271249771118164,
"eval_precision": 0.7620396600566572,
"eval_recall": 0.8213740458015267,
"eval_runtime": 2.1506,
"eval_samples_per_second": 128.336,
"eval_steps_per_second": 16.275,
"step": 4800
},
{
"epoch": 83.33,
"learning_rate": 3.3557046979865775e-05,
"loss": 0.0012,
"step": 5000
},
{
"epoch": 83.33,
"eval_accuracy_score": 0.9664105106593951,
"eval_f1": 0.777209642074507,
"eval_loss": 0.25782376527786255,
"eval_precision": 0.7450980392156863,
"eval_recall": 0.8122137404580153,
"eval_runtime": 2.1506,
"eval_samples_per_second": 128.335,
"eval_steps_per_second": 16.274,
"step": 5000
},
{
"epoch": 86.67,
"eval_accuracy_score": 0.967649975210709,
"eval_f1": 0.7958579881656805,
"eval_loss": 0.25114643573760986,
"eval_precision": 0.7718794835007173,
"eval_recall": 0.8213740458015267,
"eval_runtime": 2.1502,
"eval_samples_per_second": 128.359,
"eval_steps_per_second": 16.277,
"step": 5200
},
{
"epoch": 90.0,
"eval_accuracy_score": 0.9675260287555776,
"eval_f1": 0.7870302137067059,
"eval_loss": 0.25407665967941284,
"eval_precision": 0.7606837606837606,
"eval_recall": 0.815267175572519,
"eval_runtime": 2.1494,
"eval_samples_per_second": 128.408,
"eval_steps_per_second": 16.284,
"step": 5400
},
{
"epoch": 91.67,
"learning_rate": 3.1879194630872485e-05,
"loss": 0.0011,
"step": 5500
},
{
"epoch": 93.33,
"eval_accuracy_score": 0.9671541893901835,
"eval_f1": 0.8008817046289494,
"eval_loss": 0.2622869312763214,
"eval_precision": 0.7719546742209632,
"eval_recall": 0.8320610687022901,
"eval_runtime": 2.1439,
"eval_samples_per_second": 128.74,
"eval_steps_per_second": 16.326,
"step": 5600
},
{
"epoch": 96.67,
"eval_accuracy_score": 0.9691373326722855,
"eval_f1": 0.8053293856402666,
"eval_loss": 0.25844594836235046,
"eval_precision": 0.7816091954022989,
"eval_recall": 0.8305343511450382,
"eval_runtime": 2.1401,
"eval_samples_per_second": 128.969,
"eval_steps_per_second": 16.355,
"step": 5800
},
{
"epoch": 100.0,
"learning_rate": 3.02013422818792e-05,
"loss": 0.0013,
"step": 6000
},
{
"epoch": 100.0,
"eval_accuracy_score": 0.9682697074863659,
"eval_f1": 0.7958732498157702,
"eval_loss": 0.25458037853240967,
"eval_precision": 0.7692307692307693,
"eval_recall": 0.8244274809160306,
"eval_runtime": 2.1443,
"eval_samples_per_second": 128.714,
"eval_steps_per_second": 16.322,
"step": 6000
},
{
"epoch": 103.33,
"eval_accuracy_score": 0.9678978681209718,
"eval_f1": 0.7973273942093543,
"eval_loss": 0.25813519954681396,
"eval_precision": 0.7760115606936416,
"eval_recall": 0.8198473282442749,
"eval_runtime": 2.1415,
"eval_samples_per_second": 128.885,
"eval_steps_per_second": 16.344,
"step": 6200
},
{
"epoch": 106.67,
"eval_accuracy_score": 0.9687654933068914,
"eval_f1": 0.795234549516009,
"eval_loss": 0.2546005845069885,
"eval_precision": 0.7761627906976745,
"eval_recall": 0.815267175572519,
"eval_runtime": 2.1444,
"eval_samples_per_second": 128.705,
"eval_steps_per_second": 16.321,
"step": 6400
},
{
"epoch": 108.33,
"learning_rate": 2.8523489932885905e-05,
"loss": 0.0011,
"step": 6500
},
{
"epoch": 110.0,
"eval_accuracy_score": 0.967649975210709,
"eval_f1": 0.796711509715994,
"eval_loss": 0.2577730715274811,
"eval_precision": 0.780380673499268,
"eval_recall": 0.8137404580152672,
"eval_runtime": 2.1446,
"eval_samples_per_second": 128.696,
"eval_steps_per_second": 16.32,
"step": 6600
},
{
"epoch": 113.33,
"eval_accuracy_score": 0.9678978681209718,
"eval_f1": 0.8061901252763449,
"eval_loss": 0.24760644137859344,
"eval_precision": 0.7792022792022792,
"eval_recall": 0.8351145038167939,
"eval_runtime": 2.1452,
"eval_samples_per_second": 128.66,
"eval_steps_per_second": 16.316,
"step": 6800
},
{
"epoch": 116.67,
"learning_rate": 2.6845637583892618e-05,
"loss": 0.0007,
"step": 7000
},
{
"epoch": 116.67,
"eval_accuracy_score": 0.9688894397620228,
"eval_f1": 0.8020833333333334,
"eval_loss": 0.25953975319862366,
"eval_precision": 0.7822931785195936,
"eval_recall": 0.8229007633587786,
"eval_runtime": 2.1496,
"eval_samples_per_second": 128.395,
"eval_steps_per_second": 16.282,
"step": 7000
},
{
"epoch": 120.0,
"eval_accuracy_score": 0.9687654933068914,
"eval_f1": 0.7976278724981469,
"eval_loss": 0.2510833740234375,
"eval_precision": 0.7752161383285303,
"eval_recall": 0.8213740458015267,
"eval_runtime": 2.1454,
"eval_samples_per_second": 128.649,
"eval_steps_per_second": 16.314,
"step": 7200
},
{
"epoch": 123.33,
"eval_accuracy_score": 0.9681457610312345,
"eval_f1": 0.7857672349888806,
"eval_loss": 0.26905274391174316,
"eval_precision": 0.7636887608069164,
"eval_recall": 0.8091603053435115,
"eval_runtime": 2.1419,
"eval_samples_per_second": 128.86,
"eval_steps_per_second": 16.341,
"step": 7400
},
{
"epoch": 125.0,
"learning_rate": 2.516778523489933e-05,
"loss": 0.0006,
"step": 7500
},
{
"epoch": 126.67,
"eval_accuracy_score": 0.9682697074863659,
"eval_f1": 0.7914201183431951,
"eval_loss": 0.26672181487083435,
"eval_precision": 0.7675753228120517,
"eval_recall": 0.816793893129771,
"eval_runtime": 2.1398,
"eval_samples_per_second": 128.981,
"eval_steps_per_second": 16.356,
"step": 7600
},
{
"epoch": 130.0,
"eval_accuracy_score": 0.9686415468517601,
"eval_f1": 0.7979351032448379,
"eval_loss": 0.2618446350097656,
"eval_precision": 0.7717546362339515,
"eval_recall": 0.8259541984732824,
"eval_runtime": 2.1453,
"eval_samples_per_second": 128.65,
"eval_steps_per_second": 16.314,
"step": 7800
},
{
"epoch": 133.33,
"learning_rate": 2.348993288590604e-05,
"loss": 0.0006,
"step": 8000
},
{
"epoch": 133.33,
"eval_accuracy_score": 0.967030242935052,
"eval_f1": 0.7997086671522214,
"eval_loss": 0.27878984808921814,
"eval_precision": 0.7646239554317549,
"eval_recall": 0.8381679389312977,
"eval_runtime": 2.1418,
"eval_samples_per_second": 128.862,
"eval_steps_per_second": 16.341,
"step": 8000
},
{
"epoch": 133.33,
"step": 8000,
"total_flos": 4.186468556207309e+16,
"train_loss": 0.030014565151184797,
"train_runtime": 3978.2234,
"train_samples_per_second": 120.657,
"train_steps_per_second": 3.771
}
],
"max_steps": 15000,
"num_train_epochs": 250,
"total_flos": 4.186468556207309e+16,
"trial_name": null,
"trial_params": null
}