| { |
| "pretrained_model_name_or_path": "hfl/chinese-macbert-base", |
| "path_relm": "", |
| "path_train": "train.v15.train.json.zh_punct", |
| "path_dev": "dev.v10.dev.json.zh_punct", |
| "path_tet": "test.json.zh_punct", |
| "model_save_path": "../output/csc_merge_9_public_of_mdcspell_add_v10_extend_2000w_pinyin_init_v5_bs32_lr2e5_epoch7_recall", |
| "task_name": "csc_merge_9_public_of_mdcspell_add_v10_extend_2000w_pinyin_init_v5_bs32_lr2e5_epoch7_recall", |
| "use_fast_tokenizer": true, |
| "do_lower_case": true, |
| "do_train": true, |
| "do_eval": true, |
| "do_test": true, |
| "gradient_accumulation_steps": 4, |
| "warmup_proportion": null, |
| "num_warmup_steps": 128, |
| "max_train_steps": 1101576, |
| "num_train_epochs": 7, |
| "train_batch_size": 32, |
| "eval_batch_size": 32, |
| "learning_rate": 3e-05, |
| "max_seq_length": 128, |
| "max_grad_norm": 1.0, |
| "weight_decay": 0.0005, |
| "save_steps": 10000, |
| "anchor": null, |
| "seed": 42, |
| "lr_scheduler_type": "cosine", |
| "loss_type": "focal_loss", |
| "mask_mode": "noerror", |
| "loss_det_rate": 0.15, |
| "prompt_length": 0, |
| "mask_rate": 0.15, |
| "threshold": 0.5, |
| "flag_dynamic_encode": false, |
| "flag_loss_period": false, |
| "flag_cpo_loss": false, |
| "flag_pin_memory": true, |
| "flag_train": false, |
| "flag_fp16": false, |
| "flag_cuda": true, |
| "flag_skip": true, |
| "flag_mft": true, |
| "num_workers": 4, |
| "CUDA_VISIBLE_DEVICES": "0", |
| "USE_TORCH": "1" |
| } |