| { |
| "best_metric": 89.06060606060606, |
| "best_model_checkpoint": "/kaggle/working/whisper-bangla-dialect/checkpoint-754", |
| "epoch": 4.0, |
| "eval_steps": 500, |
| "global_step": 754, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.26525198938992045, |
| "grad_norm": 0.3513985276222229, |
| "learning_rate": 5e-06, |
| "loss": 1.4134, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.5305039787798409, |
| "grad_norm": 1.5132211446762085, |
| "learning_rate": 1e-05, |
| "loss": 1.4147, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.7957559681697612, |
| "grad_norm": 1.2995154857635498, |
| "learning_rate": 1.5000000000000002e-05, |
| "loss": 1.4236, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.9973474801061007, |
| "eval_loss": 1.489962100982666, |
| "eval_norm_levenshtein_similarity": 84.11111111111111, |
| "eval_runtime": 247.0283, |
| "eval_samples_per_second": 1.356, |
| "eval_steps_per_second": 0.34, |
| "eval_wer": 23.701657458563535, |
| "step": 188 |
| }, |
| { |
| "epoch": 1.0610079575596818, |
| "grad_norm": 1.1242467164993286, |
| "learning_rate": 2e-05, |
| "loss": 1.4245, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.3262599469496021, |
| "grad_norm": 1.6616029739379883, |
| "learning_rate": 1.9775552389476865e-05, |
| "loss": 1.4276, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.5915119363395225, |
| "grad_norm": 0.9967484474182129, |
| "learning_rate": 1.911228490388136e-05, |
| "loss": 1.4284, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.8567639257294428, |
| "grad_norm": 1.0617278814315796, |
| "learning_rate": 1.8039971303669407e-05, |
| "loss": 1.4261, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 1.4839389324188232, |
| "eval_norm_levenshtein_similarity": 86.65656565656565, |
| "eval_runtime": 239.6146, |
| "eval_samples_per_second": 1.398, |
| "eval_steps_per_second": 0.351, |
| "eval_wer": 19.723756906077348, |
| "step": 377 |
| }, |
| { |
| "epoch": 2.1220159151193636, |
| "grad_norm": 0.8240700364112854, |
| "learning_rate": 1.6606747233900816e-05, |
| "loss": 1.4236, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.387267904509284, |
| "grad_norm": 0.859289824962616, |
| "learning_rate": 1.4876949438136348e-05, |
| "loss": 1.4214, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.6525198938992043, |
| "grad_norm": 1.629287600517273, |
| "learning_rate": 1.2928227712765504e-05, |
| "loss": 1.4193, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.9177718832891246, |
| "grad_norm": 0.6796701550483704, |
| "learning_rate": 1.0848059244755093e-05, |
| "loss": 1.4182, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.9973474801061006, |
| "eval_loss": 1.4657394886016846, |
| "eval_norm_levenshtein_similarity": 88.65656565656566, |
| "eval_runtime": 244.2236, |
| "eval_samples_per_second": 1.372, |
| "eval_steps_per_second": 0.344, |
| "eval_wer": 15.91160220994475, |
| "step": 565 |
| }, |
| { |
| "epoch": 3.183023872679045, |
| "grad_norm": 1.5891698598861694, |
| "learning_rate": 8.729821802531213e-06, |
| "loss": 1.4163, |
| "step": 600 |
| }, |
| { |
| "epoch": 3.4482758620689653, |
| "grad_norm": 0.36193588376045227, |
| "learning_rate": 6.668602052579425e-06, |
| "loss": 1.4151, |
| "step": 650 |
| }, |
| { |
| "epoch": 3.713527851458886, |
| "grad_norm": 0.22459140419960022, |
| "learning_rate": 4.756927164427685e-06, |
| "loss": 1.4148, |
| "step": 700 |
| }, |
| { |
| "epoch": 3.9787798408488064, |
| "grad_norm": 0.2746105194091797, |
| "learning_rate": 3.0806113102245395e-06, |
| "loss": 1.4142, |
| "step": 750 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 1.4645167589187622, |
| "eval_norm_levenshtein_similarity": 89.06060606060606, |
| "eval_runtime": 238.9711, |
| "eval_samples_per_second": 1.402, |
| "eval_steps_per_second": 0.352, |
| "eval_wer": 14.806629834254142, |
| "step": 754 |
| } |
| ], |
| "logging_steps": 50, |
| "max_steps": 940, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.4803399278592e+18, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
