| { | |
| "best_metric": 0.3981592158737748, | |
| "best_model_checkpoint": "w2v-bert-2.0-CV_Fleurs-lg-10hrs-v4/checkpoint-5643", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 5643, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 30.792112350463867, | |
| "learning_rate": 9.998031496062993e-05, | |
| "loss": 1.4106, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.1264185884550246, | |
| "eval_loss": 0.5293365716934204, | |
| "eval_runtime": 31.0205, | |
| "eval_samples_per_second": 26.628, | |
| "eval_steps_per_second": 6.673, | |
| "eval_wer": 0.5665790102797036, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 10.16055679321289, | |
| "learning_rate": 9.897047244094488e-05, | |
| "loss": 0.4775, | |
| "step": 1026 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.11182513176846737, | |
| "eval_loss": 0.4882170259952545, | |
| "eval_runtime": 30.4398, | |
| "eval_samples_per_second": 27.135, | |
| "eval_steps_per_second": 6.8, | |
| "eval_wer": 0.5251016017212526, | |
| "step": 1026 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 24.60751724243164, | |
| "learning_rate": 9.796062992125985e-05, | |
| "loss": 0.3696, | |
| "step": 1539 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.09981164627486981, | |
| "eval_loss": 0.397329181432724, | |
| "eval_runtime": 30.5019, | |
| "eval_samples_per_second": 27.08, | |
| "eval_steps_per_second": 6.786, | |
| "eval_wer": 0.4564905570164953, | |
| "step": 1539 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 8.973501205444336, | |
| "learning_rate": 9.695078740157481e-05, | |
| "loss": 0.3007, | |
| "step": 2052 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 0.09322717991737761, | |
| "eval_loss": 0.3895888030529022, | |
| "eval_runtime": 29.8209, | |
| "eval_samples_per_second": 27.699, | |
| "eval_steps_per_second": 6.941, | |
| "eval_wer": 0.4311498924217069, | |
| "step": 2052 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.461097240447998, | |
| "learning_rate": 9.594094488188978e-05, | |
| "loss": 0.2564, | |
| "step": 2565 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 0.09659855331676664, | |
| "eval_loss": 0.3743041455745697, | |
| "eval_runtime": 29.9644, | |
| "eval_samples_per_second": 27.566, | |
| "eval_steps_per_second": 6.908, | |
| "eval_wer": 0.4372459956968683, | |
| "step": 2565 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 9.60593032836914, | |
| "learning_rate": 9.493110236220472e-05, | |
| "loss": 0.2273, | |
| "step": 3078 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.0879880973108153, | |
| "eval_loss": 0.3549309968948364, | |
| "eval_runtime": 30.2119, | |
| "eval_samples_per_second": 27.34, | |
| "eval_steps_per_second": 6.852, | |
| "eval_wer": 0.41836002868754485, | |
| "step": 3078 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.7023348808288574, | |
| "learning_rate": 9.392125984251969e-05, | |
| "loss": 0.203, | |
| "step": 3591 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.08468003608794061, | |
| "eval_loss": 0.3835783302783966, | |
| "eval_runtime": 29.6315, | |
| "eval_samples_per_second": 27.876, | |
| "eval_steps_per_second": 6.986, | |
| "eval_wer": 0.4000717188620607, | |
| "step": 3591 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.196669340133667, | |
| "learning_rate": 9.291141732283465e-05, | |
| "loss": 0.1775, | |
| "step": 4104 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.08550309438262714, | |
| "eval_loss": 0.3693934977054596, | |
| "eval_runtime": 30.2095, | |
| "eval_samples_per_second": 27.342, | |
| "eval_steps_per_second": 6.852, | |
| "eval_wer": 0.4132201769065264, | |
| "step": 4104 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 3.045036792755127, | |
| "learning_rate": 9.190354330708662e-05, | |
| "loss": 0.159, | |
| "step": 4617 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.0889694360467877, | |
| "eval_loss": 0.3487217128276825, | |
| "eval_runtime": 29.9136, | |
| "eval_samples_per_second": 27.613, | |
| "eval_steps_per_second": 6.92, | |
| "eval_wer": 0.4136983026535979, | |
| "step": 4617 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.3439998626708984, | |
| "learning_rate": 9.089370078740158e-05, | |
| "loss": 0.1444, | |
| "step": 5130 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.08915937257633075, | |
| "eval_loss": 0.3705226480960846, | |
| "eval_runtime": 30.2455, | |
| "eval_samples_per_second": 27.31, | |
| "eval_steps_per_second": 6.844, | |
| "eval_wer": 0.42768348075543866, | |
| "step": 5130 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 3.339082956314087, | |
| "learning_rate": 8.988385826771653e-05, | |
| "loss": 0.1262, | |
| "step": 5643 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.08284398296902452, | |
| "eval_loss": 0.3619905114173889, | |
| "eval_runtime": 30.7416, | |
| "eval_samples_per_second": 26.869, | |
| "eval_steps_per_second": 6.734, | |
| "eval_wer": 0.3981592158737748, | |
| "step": 5643 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 51300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 20, | |
| "early_stopping_threshold": 0.001 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.2003321769803633e+19, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |