CONFIG ├── data │ └── _target_: src.data.kaldi_pretraining_dataset.build_kaldi_datamodule │ train_split: train │ dev_splits: │ - dev_1k │ - dev_gmuaccent │ - dev_buckeye │ - dev_epadb │ - dev_speechoceanotth │ - dev_l2arctic │ dataset_config_path: configs/data/ipapack_index.yaml │ vocab_file: src/model/xeusphoneme/resources/ipa_vocab.json │ batch_size: 12 │ num_workers: 6 │ ├── model │ └── _target_: src.recipe.phone_recognition.model_module.PhoneRecognitionMode │ optimizer: │ _target_: torch.optim.Adam │ _partial_: true │ lr: 3.0e-05 │ weight_decay: 0.0 │ scheduler: │ _target_: src.recipe.phone_recognition.model_module.get_w2v2ph_schedul │ _partial_: true │ num_training_steps: 40000 │ encoder_unfreeze_step: 3000 │ warmup_fraction: 0.05 │ constant_fraction: 0.75 │ net: │ _target_: src.model.xeusphoneme.builders.build_xeus_pr_from_hf │ work_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/cache/xeus │ hf_repo: espnet/xeus │ force: false │ config_file: null │ checkpoint: null │ vocab_file: src/model/xeusphoneme/resources/ipa_vocab.json │ load_ckpt: true │ interctc_weight: 0.3 │ interctc_layer_idx: │ - 4 │ - 8 │ - 12 │ interctc_use_conditioning: true │ ctc_weight: 1.0 │ decoder_config: null │ ctc_config: │ ctc_type: builtin │ dev_splits: │ - dev_1k │ - dev_gmuaccent │ - dev_buckeye │ - dev_epadb │ - dev_speechoceanotth │ - dev_l2arctic │ ├── callbacks │ └── model_checkpoint: │ _target_: lightning.pytorch.callbacks.ModelCheckpoint │ dirpath: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/runs/ipaaccent_ │ filename: checkpoint-{step} │ monitor: null │ verbose: false │ save_last: true │ save_top_k: -1 │ mode: min │ auto_insert_metric_name: false │ save_weights_only: false │ every_n_train_steps: 2000 │ train_time_interval: null │ every_n_epochs: null │ save_on_train_epoch_end: null │ learning_rate_monitor: │ _target_: lightning.pytorch.callbacks.LearningRateMonitor │ logging_interval: step │ early_stopping: null │ model_summary: │ _target_: lightning.pytorch.callbacks.RichModelSummary │ max_depth: 1 │ rich_progress_bar: │ _target_: lightning.pytorch.callbacks.RichProgressBar │ ├── logger │ └── wandb: │ _target_: lightning.pytorch.loggers.wandb.WandbLogger │ save_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/runs/ipaaccent │ offline: false │ id: null │ anonymous: null │ project: SSLPR │ log_model: false │ prefix: '' │ group: ipaaccent_ctc │ tags: │ - ipapack_accent │ - ctc │ - phone_recognition │ job_type: '' │ name: xeus_multiaccent.selfctc_l4_8_12.bs256.lr3em5.sched_p05warm_p75c │ ├── trainer │ └── _target_: lightning.pytorch.trainer.Trainer │ default_root_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/runs/ipa │ min_epochs: 0 │ max_epochs: -1 │ accelerator: gpu │ devices: auto │ check_val_every_n_epoch: null │ deterministic: false │ num_nodes: 1 │ strategy: ddp_find_unused_parameters_true │ precision: bf16 │ sync_batchnorm: true │ max_steps: 40000 │ gradient_clip_val: 1.0 │ accumulate_grad_batches: 5 │ val_check_interval: 2000 │ log_every_n_steps: 200 │ ├── paths │ └── root_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench │ exp_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp │ download_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/download │ cache_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/cache │ output_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench/exp/runs/ipaaccent │ work_dir: /mnt/home/chinjoul-andr-daf644/PhoneBench │ ├── extras │ └── ignore_warnings: false │ enforce_tags: true │ print_config: true │ ├── task_name │ └── ipaaccent_ctc ├── tags │ └── ['ipapack_accent', 'ctc', 'phone_recognition'] ├── train │ └── True ├── test │ └── True ├── predict │ └── False ├── pred_file │ └── None ├── run_folder │ └── xeus_multiaccent.selfctc_l4_8_12.bs256.lr3em5.sched_p05warm_p75const_3ku ├── ckpt_path │ └── None └── seed └── 42