| { |
| "best_global_step": 2600, |
| "best_metric": 0.5833333333333334, |
| "best_model_checkpoint": "/kaggle/working/wav2vec2-bengali-finetuned/checkpoint-2400", |
| "epoch": 30.0, |
| "eval_steps": 200, |
| "global_step": 2940, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.01020408163265306, |
| "grad_norm": 1263584.0, |
| "learning_rate": 0.0, |
| "loss": 25.6736, |
| "step": 1 |
| }, |
| { |
| "epoch": 1.0204081632653061, |
| "grad_norm": 3843419.25, |
| "learning_rate": 9.900000000000002e-06, |
| "loss": 23.811, |
| "step": 100 |
| }, |
| { |
| "epoch": 2.0408163265306123, |
| "grad_norm": 1233849.75, |
| "learning_rate": 1.9900000000000003e-05, |
| "loss": 13.9598, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.0408163265306123, |
| "eval_loss": 5.328553199768066, |
| "eval_runtime": 11.0357, |
| "eval_samples_per_second": 14.951, |
| "eval_steps_per_second": 1.903, |
| "eval_wer": 1.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.061224489795918, |
| "grad_norm": 101215.4375, |
| "learning_rate": 2.9900000000000002e-05, |
| "loss": 4.6042, |
| "step": 300 |
| }, |
| { |
| "epoch": 4.081632653061225, |
| "grad_norm": 204473.296875, |
| "learning_rate": 3.99e-05, |
| "loss": 3.5131, |
| "step": 400 |
| }, |
| { |
| "epoch": 4.081632653061225, |
| "eval_loss": 3.380544900894165, |
| "eval_runtime": 10.5068, |
| "eval_samples_per_second": 15.704, |
| "eval_steps_per_second": 1.999, |
| "eval_wer": 1.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 5.1020408163265305, |
| "grad_norm": 106750.3828125, |
| "learning_rate": 4.99e-05, |
| "loss": 3.2448, |
| "step": 500 |
| }, |
| { |
| "epoch": 6.122448979591836, |
| "grad_norm": 56899.28125, |
| "learning_rate": 4.7971311475409835e-05, |
| "loss": 2.9613, |
| "step": 600 |
| }, |
| { |
| "epoch": 6.122448979591836, |
| "eval_loss": 2.839696168899536, |
| "eval_runtime": 10.5616, |
| "eval_samples_per_second": 15.623, |
| "eval_steps_per_second": 1.988, |
| "eval_wer": 1.0, |
| "step": 600 |
| }, |
| { |
| "epoch": 7.142857142857143, |
| "grad_norm": 155864.65625, |
| "learning_rate": 4.5922131147540986e-05, |
| "loss": 2.8199, |
| "step": 700 |
| }, |
| { |
| "epoch": 8.16326530612245, |
| "grad_norm": 75451.0546875, |
| "learning_rate": 4.387295081967213e-05, |
| "loss": 2.7197, |
| "step": 800 |
| }, |
| { |
| "epoch": 8.16326530612245, |
| "eval_loss": 2.606015920639038, |
| "eval_runtime": 10.535, |
| "eval_samples_per_second": 15.662, |
| "eval_steps_per_second": 1.993, |
| "eval_wer": 0.8457207207207207, |
| "step": 800 |
| }, |
| { |
| "epoch": 9.183673469387756, |
| "grad_norm": 76782.7421875, |
| "learning_rate": 4.182377049180328e-05, |
| "loss": 2.6389, |
| "step": 900 |
| }, |
| { |
| "epoch": 10.204081632653061, |
| "grad_norm": 64796.6171875, |
| "learning_rate": 3.977459016393443e-05, |
| "loss": 2.5596, |
| "step": 1000 |
| }, |
| { |
| "epoch": 10.204081632653061, |
| "eval_loss": 2.4175667762756348, |
| "eval_runtime": 10.5336, |
| "eval_samples_per_second": 15.664, |
| "eval_steps_per_second": 1.994, |
| "eval_wer": 0.7815315315315315, |
| "step": 1000 |
| }, |
| { |
| "epoch": 11.224489795918368, |
| "grad_norm": 147111.484375, |
| "learning_rate": 3.772540983606558e-05, |
| "loss": 2.4538, |
| "step": 1100 |
| }, |
| { |
| "epoch": 12.244897959183673, |
| "grad_norm": 91618.7890625, |
| "learning_rate": 3.5676229508196724e-05, |
| "loss": 2.3313, |
| "step": 1200 |
| }, |
| { |
| "epoch": 12.244897959183673, |
| "eval_loss": 2.050769805908203, |
| "eval_runtime": 10.5394, |
| "eval_samples_per_second": 15.656, |
| "eval_steps_per_second": 1.993, |
| "eval_wer": 0.7792792792792793, |
| "step": 1200 |
| }, |
| { |
| "epoch": 13.26530612244898, |
| "grad_norm": 135995.46875, |
| "learning_rate": 3.362704918032787e-05, |
| "loss": 2.1482, |
| "step": 1300 |
| }, |
| { |
| "epoch": 14.285714285714286, |
| "grad_norm": 91693.046875, |
| "learning_rate": 3.157786885245902e-05, |
| "loss": 1.9715, |
| "step": 1400 |
| }, |
| { |
| "epoch": 14.285714285714286, |
| "eval_loss": 1.6378555297851562, |
| "eval_runtime": 10.5414, |
| "eval_samples_per_second": 15.653, |
| "eval_steps_per_second": 1.992, |
| "eval_wer": 0.7894144144144144, |
| "step": 1400 |
| }, |
| { |
| "epoch": 15.306122448979592, |
| "grad_norm": 106778.4609375, |
| "learning_rate": 2.9528688524590165e-05, |
| "loss": 1.8192, |
| "step": 1500 |
| }, |
| { |
| "epoch": 16.3265306122449, |
| "grad_norm": 93321.4921875, |
| "learning_rate": 2.7479508196721317e-05, |
| "loss": 1.706, |
| "step": 1600 |
| }, |
| { |
| "epoch": 16.3265306122449, |
| "eval_loss": 1.3998359441757202, |
| "eval_runtime": 10.5682, |
| "eval_samples_per_second": 15.613, |
| "eval_steps_per_second": 1.987, |
| "eval_wer": 0.7263513513513513, |
| "step": 1600 |
| }, |
| { |
| "epoch": 17.346938775510203, |
| "grad_norm": 111310.1328125, |
| "learning_rate": 2.5430327868852462e-05, |
| "loss": 1.5935, |
| "step": 1700 |
| }, |
| { |
| "epoch": 18.367346938775512, |
| "grad_norm": 131049.40625, |
| "learning_rate": 2.338114754098361e-05, |
| "loss": 1.5412, |
| "step": 1800 |
| }, |
| { |
| "epoch": 18.367346938775512, |
| "eval_loss": 1.2491974830627441, |
| "eval_runtime": 10.5527, |
| "eval_samples_per_second": 15.636, |
| "eval_steps_per_second": 1.99, |
| "eval_wer": 0.6722972972972973, |
| "step": 1800 |
| }, |
| { |
| "epoch": 19.387755102040817, |
| "grad_norm": 150843.8125, |
| "learning_rate": 2.1331967213114755e-05, |
| "loss": 1.4727, |
| "step": 1900 |
| }, |
| { |
| "epoch": 20.408163265306122, |
| "grad_norm": 124916.6015625, |
| "learning_rate": 1.9282786885245903e-05, |
| "loss": 1.4023, |
| "step": 2000 |
| }, |
| { |
| "epoch": 20.408163265306122, |
| "eval_loss": 1.138938546180725, |
| "eval_runtime": 10.5847, |
| "eval_samples_per_second": 15.589, |
| "eval_steps_per_second": 1.984, |
| "eval_wer": 0.6317567567567568, |
| "step": 2000 |
| }, |
| { |
| "epoch": 21.428571428571427, |
| "grad_norm": 260327.703125, |
| "learning_rate": 1.7233606557377048e-05, |
| "loss": 1.3585, |
| "step": 2100 |
| }, |
| { |
| "epoch": 22.448979591836736, |
| "grad_norm": 156561.15625, |
| "learning_rate": 1.5184426229508198e-05, |
| "loss": 1.3188, |
| "step": 2200 |
| }, |
| { |
| "epoch": 22.448979591836736, |
| "eval_loss": 1.0857497453689575, |
| "eval_runtime": 10.5312, |
| "eval_samples_per_second": 15.668, |
| "eval_steps_per_second": 1.994, |
| "eval_wer": 0.6148648648648649, |
| "step": 2200 |
| }, |
| { |
| "epoch": 23.46938775510204, |
| "grad_norm": 137688.609375, |
| "learning_rate": 1.3135245901639346e-05, |
| "loss": 1.2872, |
| "step": 2300 |
| }, |
| { |
| "epoch": 24.489795918367346, |
| "grad_norm": 149320.140625, |
| "learning_rate": 1.1086065573770492e-05, |
| "loss": 1.2707, |
| "step": 2400 |
| }, |
| { |
| "epoch": 24.489795918367346, |
| "eval_loss": 1.035390019416809, |
| "eval_runtime": 10.5453, |
| "eval_samples_per_second": 15.647, |
| "eval_steps_per_second": 1.991, |
| "eval_wer": 0.6047297297297297, |
| "step": 2400 |
| }, |
| { |
| "epoch": 25.510204081632654, |
| "grad_norm": 160687.875, |
| "learning_rate": 9.036885245901639e-06, |
| "loss": 1.2261, |
| "step": 2500 |
| }, |
| { |
| "epoch": 26.53061224489796, |
| "grad_norm": 175958.0625, |
| "learning_rate": 6.987704918032788e-06, |
| "loss": 1.222, |
| "step": 2600 |
| }, |
| { |
| "epoch": 26.53061224489796, |
| "eval_loss": 1.011266827583313, |
| "eval_runtime": 10.5487, |
| "eval_samples_per_second": 15.642, |
| "eval_steps_per_second": 1.991, |
| "eval_wer": 0.5833333333333334, |
| "step": 2600 |
| }, |
| { |
| "epoch": 27.551020408163264, |
| "grad_norm": 129829.171875, |
| "learning_rate": 4.938524590163935e-06, |
| "loss": 1.2061, |
| "step": 2700 |
| }, |
| { |
| "epoch": 28.571428571428573, |
| "grad_norm": 180224.375, |
| "learning_rate": 2.8893442622950824e-06, |
| "loss": 1.2084, |
| "step": 2800 |
| }, |
| { |
| "epoch": 28.571428571428573, |
| "eval_loss": 0.9903889298439026, |
| "eval_runtime": 10.5794, |
| "eval_samples_per_second": 15.596, |
| "eval_steps_per_second": 1.985, |
| "eval_wer": 0.5833333333333334, |
| "step": 2800 |
| }, |
| { |
| "epoch": 29.591836734693878, |
| "grad_norm": 153736.046875, |
| "learning_rate": 8.401639344262295e-07, |
| "loss": 1.1881, |
| "step": 2900 |
| }, |
| { |
| "epoch": 30.0, |
| "step": 2940, |
| "total_flos": 1.8456230162149265e+19, |
| "train_loss": 3.165084928395797, |
| "train_runtime": 15598.205, |
| "train_samples_per_second": 6.024, |
| "train_steps_per_second": 0.188 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 2940, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 30, |
| "save_steps": 400, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 5, |
| "early_stopping_threshold": 0.01 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 1 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.8456230162149265e+19, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
