update csc t5 model

README.md

@@ -1,3 +1,31 @@
 ---
 license: apache-2.0
 ---
+
+# CSC T5 - T5 for Traditional Chinese Spelling Correction
+
+This model was obtained by `instruction-tuning` the corresponding `ClueAI/PromptCLUE-base-v1-5` model on the spelling error corpus. 
+
+## Model Details
+### Model Description
+- Language(s) (NLP): `Chinese`
+- Pretrained from model: `ClueAI/PromptCLUE-base-v1-5`
+- Pretrained by dataset: `1M UDN news corpus`
+- Finetuned by dataset: `shibing624/CSC` spelling error corpus
+
+### Model Sources
+- Repository: [https://github.com/TedYeh/Chinese_spelling_Correction](https://github.com/TedYeh/Chinese_spelling_Correction)
+
+## Usage
+```python
+from transformers import AutoTokenizer, T5ForConditionalGeneration
+tokenizer = AutoTokenizer.from_pretrained("CodeTed/traditional_CSC_t5")
+model = T5ForConditionalGeneration.from_pretrained("CodeTed/traditional_CSC_t5")
+input_text = '糾正句子裡的錯字: 為了降低少子化，政府可以堆動獎勵生育的政策。'
+input_ids = tokenizer(input_text, return_tensors="pt").input_ids
+outputs = model.generate(input_ids, max_length=256)
+edited_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
+```
+
+### Related Project
+[CodeTed/CGEDit](https://huggingface.co/CodeTed/CGEDit)

config.json

@@ -0,0 +1,31 @@
+{
+  "_name_or_path": "./outputs/prompt_1m",
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "d_ff": 2048,
+  "d_kv": 64,
+  "d_model": 768,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
+  "dropout_rate": 0.1,
+  "eos_token_id": 1,
+  "feed_forward_proj": "gated-gelu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "num_decoder_layers": 12,
+  "num_heads": 12,
+  "num_layers": 12,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float32",
+  "transformers_version": "4.30.2",
+  "use_cache": true,
+  "vocab_size": 40043
+}

eval_results.txt

@@ -0,0 +1 @@
+eval_loss = 0.03081325274493281

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "decoder_start_token_id": 0,
+  "eos_token_id": 1,
+  "pad_token_id": 0,
+  "transformers_version": "4.30.2"
+}

model_args.json

@@ -0,0 +1 @@
+{"adafactor_beta1": null, "adafactor_clip_threshold": 1.0, "adafactor_decay_rate": -0.8, "adafactor_eps": [1e-30, 0.001], "adafactor_relative_step": false, "adafactor_scale_parameter": false, "adafactor_warmup_init": false, "adam_epsilon": 1e-08, "best_model_dir": "./outputs/prompt_pretrain_271k_trad_sim/best_model", "cache_dir": "cache_dir/", "config": {}, "cosine_schedule_num_cycles": 0.5, "custom_layer_parameters": [], "custom_parameter_groups": [], "dataloader_num_workers": 0, "do_lower_case": false, "dynamic_quantize": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "eval_loss", "early_stopping_metric_minimize": true, "early_stopping_patience": 3, "encoding": "utf-8", "eval_batch_size": 8, "evaluate_during_training": true, "evaluate_during_training_silent": true, "evaluate_during_training_steps": 6000, "evaluate_during_training_verbose": true, "evaluate_each_epoch": true, "fp16": false, "gradient_accumulation_steps": 1, "learning_rate": 0.0005, "local_rank": -1, "logging_steps": 50, "manual_seed": null, "max_grad_norm": 1.0, "max_seq_length": 200, "model_name": "./outputs/prompt_1m", "model_type": "t5", "multiprocessing_chunksize": -1, "n_gpu": 1, "no_cache": false, "no_save": false, "not_saved_args": [], "num_train_epochs": 20, "optimizer": "AdamW", "output_dir": "./outputs/prompt_pretrain_271k_trad_sim", "overwrite_output_dir": true, "polynomial_decay_schedule_lr_end": 1e-07, "polynomial_decay_schedule_power": 1.0, "process_count": 46, "quantized_model": false, "reprocess_input_data": true, "save_best_model": true, "save_eval_checkpoints": false, "save_model_every_epoch": true, "save_optimizer_and_scheduler": true, "save_steps": 6000, "scheduler": "constant_schedule_with_warmup", "silent": false, "skip_special_tokens": true, "tensorboard_dir": null, "thread_count": null, "tokenizer_name": null, "tokenizer_type": null, "train_batch_size": 40, "train_custom_parameters_only": false, "use_cached_eval_features": false, "use_early_stopping": true, "use_hf_datasets": false, "use_multiprocessing": false, "use_multiprocessing_for_evaluation": false, "wandb_kwargs": {}, "wandb_project": null, "warmup_ratio": 0.06, "warmup_steps": 17847, "weight_decay": 0.0, "model_class": "T5Model", "dataset_class": null, "do_sample": false, "early_stopping": true, "evaluate_generated_text": true, "length_penalty": 2.0, "max_length": 400, "max_steps": -1, "num_beams": 1, "num_return_sequences": 1, "preprocess_inputs": true, "repetition_penalty": 1.0, "special_tokens_list": [], "top_k": null, "top_p": null, "use_multiprocessed_decoding": false}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0375c340cb0a0c0767396cc300501bc0f0ba457b9e86624dcd5d089f79801613
+size 1039038645

special_tokens_map.json

@@ -0,0 +1,5 @@
+{
+  "eos_token": "</s>",
+  "pad_token": "<pad>",
+  "unk_token": "<unk>"
+}

spiece.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c2083a4dd8b56e99db4a408d68a50d4374252628247dbdc9d5931fddcf0a66d
+size 742045

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "additional_special_tokens": null,
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "</s>",
+  "extra_ids": 0,
+  "model_max_length": 512,
+  "pad_token": "<pad>",
+  "sp_model_kwargs": {},
+  "tokenizer_class": "T5Tokenizer",
+  "truncate": true,
+  "unk_token": "<unk>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4004c67f6811d29f97761871e890fea8561fc337cb140ef4994ebebb32edeaf
+size 3195