piplustheta's picture
Training in progress, epoch 0
dd85623
{
"best_metric": 0.804093567251462,
"best_model_checkpoint": "videomae-base-finetuned-ssv2-finetuned-samsung-subset/checkpoint-408",
"epoch": 3.2426470588235294,
"eval_steps": 500,
"global_step": 408,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 1.2195121951219513e-05,
"loss": 3.0442,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 2.4390243902439026e-05,
"loss": 2.9549,
"step": 20
},
{
"epoch": 0.07,
"learning_rate": 3.6585365853658535e-05,
"loss": 2.8135,
"step": 30
},
{
"epoch": 0.1,
"learning_rate": 4.878048780487805e-05,
"loss": 2.6622,
"step": 40
},
{
"epoch": 0.12,
"learning_rate": 4.877384196185286e-05,
"loss": 2.4479,
"step": 50
},
{
"epoch": 0.15,
"learning_rate": 4.741144414168938e-05,
"loss": 2.282,
"step": 60
},
{
"epoch": 0.17,
"learning_rate": 4.604904632152589e-05,
"loss": 1.9786,
"step": 70
},
{
"epoch": 0.2,
"learning_rate": 4.46866485013624e-05,
"loss": 1.9135,
"step": 80
},
{
"epoch": 0.22,
"learning_rate": 4.332425068119891e-05,
"loss": 1.6138,
"step": 90
},
{
"epoch": 0.25,
"learning_rate": 4.196185286103542e-05,
"loss": 1.608,
"step": 100
},
{
"epoch": 0.25,
"eval_accuracy": 0.5497076023391813,
"eval_loss": 1.5651301145553589,
"eval_runtime": 3380.9023,
"eval_samples_per_second": 0.303,
"eval_steps_per_second": 0.038,
"step": 103
},
{
"epoch": 1.02,
"learning_rate": 4.0599455040871935e-05,
"loss": 1.3634,
"step": 110
},
{
"epoch": 1.04,
"learning_rate": 3.923705722070845e-05,
"loss": 1.3086,
"step": 120
},
{
"epoch": 1.07,
"learning_rate": 3.787465940054496e-05,
"loss": 1.1414,
"step": 130
},
{
"epoch": 1.09,
"learning_rate": 3.651226158038147e-05,
"loss": 1.0196,
"step": 140
},
{
"epoch": 1.12,
"learning_rate": 3.5149863760217985e-05,
"loss": 0.9671,
"step": 150
},
{
"epoch": 1.14,
"learning_rate": 3.37874659400545e-05,
"loss": 0.9137,
"step": 160
},
{
"epoch": 1.16,
"learning_rate": 3.242506811989101e-05,
"loss": 0.9068,
"step": 170
},
{
"epoch": 1.19,
"learning_rate": 3.106267029972752e-05,
"loss": 1.0022,
"step": 180
},
{
"epoch": 1.21,
"learning_rate": 2.9700272479564035e-05,
"loss": 0.8898,
"step": 190
},
{
"epoch": 1.24,
"learning_rate": 2.8337874659400547e-05,
"loss": 0.7421,
"step": 200
},
{
"epoch": 1.25,
"eval_accuracy": 0.7270955165692008,
"eval_loss": 0.9049609303474426,
"eval_runtime": 3202.4383,
"eval_samples_per_second": 0.32,
"eval_steps_per_second": 0.04,
"step": 206
},
{
"epoch": 2.01,
"learning_rate": 2.6975476839237056e-05,
"loss": 0.7335,
"step": 210
},
{
"epoch": 2.03,
"learning_rate": 2.5613079019073572e-05,
"loss": 0.7486,
"step": 220
},
{
"epoch": 2.06,
"learning_rate": 2.425068119891008e-05,
"loss": 0.6743,
"step": 230
},
{
"epoch": 2.08,
"learning_rate": 2.2888283378746594e-05,
"loss": 0.5636,
"step": 240
},
{
"epoch": 2.11,
"learning_rate": 2.1525885558583106e-05,
"loss": 0.6737,
"step": 250
},
{
"epoch": 2.13,
"learning_rate": 2.016348773841962e-05,
"loss": 0.465,
"step": 260
},
{
"epoch": 2.16,
"learning_rate": 1.880108991825613e-05,
"loss": 0.6182,
"step": 270
},
{
"epoch": 2.18,
"learning_rate": 1.7438692098092644e-05,
"loss": 0.5307,
"step": 280
},
{
"epoch": 2.21,
"learning_rate": 1.6076294277929156e-05,
"loss": 0.503,
"step": 290
},
{
"epoch": 2.23,
"learning_rate": 1.4713896457765669e-05,
"loss": 0.6384,
"step": 300
},
{
"epoch": 2.25,
"eval_accuracy": 0.7962962962962963,
"eval_loss": 0.6470026969909668,
"eval_runtime": 3136.7537,
"eval_samples_per_second": 0.327,
"eval_steps_per_second": 0.041,
"step": 309
},
{
"epoch": 3.0,
"learning_rate": 1.3351498637602181e-05,
"loss": 0.4937,
"step": 310
},
{
"epoch": 3.03,
"learning_rate": 1.1989100817438692e-05,
"loss": 0.3847,
"step": 320
},
{
"epoch": 3.05,
"learning_rate": 1.0626702997275204e-05,
"loss": 0.3882,
"step": 330
},
{
"epoch": 3.08,
"learning_rate": 9.264305177111717e-06,
"loss": 0.45,
"step": 340
},
{
"epoch": 3.1,
"learning_rate": 7.90190735694823e-06,
"loss": 0.3976,
"step": 350
},
{
"epoch": 3.12,
"learning_rate": 6.539509536784741e-06,
"loss": 0.4559,
"step": 360
},
{
"epoch": 3.15,
"learning_rate": 5.1771117166212534e-06,
"loss": 0.4666,
"step": 370
},
{
"epoch": 3.17,
"learning_rate": 3.814713896457766e-06,
"loss": 0.38,
"step": 380
},
{
"epoch": 3.2,
"learning_rate": 2.452316076294278e-06,
"loss": 0.3852,
"step": 390
},
{
"epoch": 3.22,
"learning_rate": 1.0899182561307902e-06,
"loss": 0.3902,
"step": 400
},
{
"epoch": 3.24,
"eval_accuracy": 0.804093567251462,
"eval_loss": 0.5762231945991516,
"eval_runtime": 2421.5515,
"eval_samples_per_second": 0.424,
"eval_steps_per_second": 0.053,
"step": 408
},
{
"epoch": 3.24,
"step": 408,
"total_flos": 4.056591324738355e+18,
"train_loss": 1.0837838474441976,
"train_runtime": 27835.4406,
"train_samples_per_second": 0.117,
"train_steps_per_second": 0.015
},
{
"epoch": 3.24,
"eval_accuracy": 0.8138686131386861,
"eval_loss": 0.6085803508758545,
"eval_runtime": 2468.738,
"eval_samples_per_second": 0.444,
"eval_steps_per_second": 0.055,
"step": 408
}
],
"logging_steps": 10,
"max_steps": 408,
"num_train_epochs": 9223372036854775807,
"save_steps": 500,
"total_flos": 4.056591324738355e+18,
"trial_name": null,
"trial_params": null
}