| { | |
| "best_metric": 0.687998384980317, | |
| "best_model_checkpoint": "/workspace/data2/el/wav2vec2-large-xlsr-greek/checkpoint-580", | |
| "epoch": 4.9957081545064375, | |
| "global_step": 580, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.00015, | |
| "loss": 18.1246, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 0.0003, | |
| "loss": 16.1703, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 0.0002946428571428571, | |
| "loss": 6.2477, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 0.00028928571428571425, | |
| "loss": 3.5827, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 0.00028392857142857137, | |
| "loss": 3.2333, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 0.00027857142857142854, | |
| "loss": 3.1704, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0002732142857142857, | |
| "loss": 3.1395, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.00026785714285714287, | |
| "loss": 3.1397, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0002625, | |
| "loss": 3.1298, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 0.0002571428571428571, | |
| "loss": 3.1082, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 0.00025178571428571426, | |
| "loss": 3.0917, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.068535089492798, | |
| "eval_runtime": 645.4744, | |
| "eval_samples_per_second": 2.358, | |
| "eval_wer": 1.0, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.0002464285714285714, | |
| "loss": 3.2085, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.00024107142857142857, | |
| "loss": 3.0118, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "learning_rate": 0.00023571428571428569, | |
| "loss": 3.0139, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 0.00023035714285714285, | |
| "loss": 3.0176, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.000225, | |
| "loss": 2.9979, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0002196428571428571, | |
| "loss": 2.9721, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.00021428571428571427, | |
| "loss": 2.9585, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "learning_rate": 0.0002089285714285714, | |
| "loss": 2.9578, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 0.00020357142857142856, | |
| "loss": 2.9347, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0001982142857142857, | |
| "loss": 2.9348, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.00019285714285714286, | |
| "loss": 2.9156, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.00018749999999999998, | |
| "loss": 2.8911, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.9123377799987793, | |
| "eval_runtime": 651.1025, | |
| "eval_samples_per_second": 2.338, | |
| "eval_wer": 1.0, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.00018214285714285712, | |
| "loss": 3.002, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00017678571428571428, | |
| "loss": 2.7696, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 0.0001714285714285714, | |
| "loss": 2.6308, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 0.00016607142857142857, | |
| "loss": 2.3949, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 0.0001607142857142857, | |
| "loss": 2.0719, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.00015535714285714285, | |
| "loss": 1.7915, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.00015, | |
| "loss": 1.5479, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "learning_rate": 0.00014464285714285713, | |
| "loss": 1.4167, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.76, | |
| "learning_rate": 0.00013928571428571427, | |
| "loss": 1.2183, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00013392857142857144, | |
| "loss": 1.1713, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 0.00012857142857142855, | |
| "loss": 1.0325, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8776374459266663, | |
| "eval_runtime": 651.452, | |
| "eval_samples_per_second": 2.336, | |
| "eval_wer": 0.8105380034319168, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "learning_rate": 0.0001232142857142857, | |
| "loss": 0.9978, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.00011785714285714284, | |
| "loss": 0.8941, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.19, | |
| "learning_rate": 0.0001125, | |
| "loss": 0.875, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.27, | |
| "learning_rate": 0.00010714285714285714, | |
| "loss": 0.801, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "learning_rate": 0.00010178571428571428, | |
| "loss": 0.7847, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.45, | |
| "learning_rate": 9.642857142857143e-05, | |
| "loss": 0.6949, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 9.107142857142856e-05, | |
| "loss": 0.7307, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.62, | |
| "learning_rate": 8.57142857142857e-05, | |
| "loss": 0.6729, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "learning_rate": 8.035714285714285e-05, | |
| "loss": 0.6699, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.79, | |
| "learning_rate": 7.5e-05, | |
| "loss": 0.6768, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 6.964285714285713e-05, | |
| "loss": 0.6459, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 6.428571428571427e-05, | |
| "loss": 0.6589, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6442593336105347, | |
| "eval_runtime": 653.4327, | |
| "eval_samples_per_second": 2.329, | |
| "eval_wer": 0.7133340062582013, | |
| "step": 464 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 5.892857142857142e-05, | |
| "loss": 0.6647, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 5.357142857142857e-05, | |
| "loss": 0.6859, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "learning_rate": 4.8214285714285716e-05, | |
| "loss": 0.6298, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 4.285714285714285e-05, | |
| "loss": 0.6076, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 3.75e-05, | |
| "loss": 0.5485, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 3.214285714285714e-05, | |
| "loss": 0.5852, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "learning_rate": 2.6785714285714284e-05, | |
| "loss": 0.5452, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 2.1428571428571425e-05, | |
| "loss": 0.5593, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 1.607142857142857e-05, | |
| "loss": 0.5316, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 1.0714285714285712e-05, | |
| "loss": 0.5742, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 5.357142857142856e-06, | |
| "loss": 0.5978, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.5398, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.6011346578598022, | |
| "eval_runtime": 660.5778, | |
| "eval_samples_per_second": 2.304, | |
| "eval_wer": 0.687998384980317, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 580, | |
| "total_flos": 3.7562509460378655e+18, | |
| "train_runtime": 11825.7565, | |
| "train_samples_per_second": 0.049 | |
| } | |
| ], | |
| "max_steps": 580, | |
| "num_train_epochs": 5, | |
| "total_flos": 3.7562509460378655e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |