Initialize README file

README.md

@@ -0,0 +1,94 @@
+---
+language:
+- zh
+tags:
+- Seq2SeqLM
+- 古文
+- 文言文
+- 中国古代官职地名拆分
+- ancient
+- classical
+license: cc-by-nc-sa-4.0
+---
+
+# <font color="IndianRed"> OTAS (Office Title Address Splitter)</font>
+[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1UoG3QebyBlK6diiYckiQv-5dRB9dA4iv?usp=sharing/)
+
+Our model <font color="cornflowerblue">OTAS (Office Title Address Splitter) </font> is a Named Entity Recognition Classical Chinese language model that is intended to  <font color="IndianRed">split the address portion in Classical Chinese office titles.</font>. This model is first inherited from raynardj/classical-chinese-punctuation-guwen-biaodian Classical Chinese punctuation model, and finetuned using over a 25,000 high-quality punctuation pairs collected CBDB group (China Biographical Database). 
+
+### <font color="IndianRed"> How to use </font>
+
+Here is how to use this model to get the features of a given text in PyTorch:
+
+<font color="cornflowerblue"> 1. Import model and packages </font>
+```python
+from transformers import AutoTokenizer, AutoModelForTokenClassification
+
+device = torch.device('cuda')
+model_name = 'cbdb/OfficeTitleAddressSplitter'
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForTokenClassification.from_pretrained(model_name).to(device)
+```
+
+<font color="cornflowerblue"> 2. Load Data </font>
+```python
+# Load your data here
+tobe_splitted = ['湖南常德協中軍都司','廣東鹽運使','漢軍鑲黃旗副都統']
+```
+
+work-in-progress
+
+<font color="cornflowerblue"> 3. Make a prediction </font>
+```python
+
+  tokens_test = tokenizer.encode_plus(
+      tobe_splitted,
+      add_special_tokens=True,
+      return_attention_mask=True,
+      padding=True,
+      max_length=max_seq_len,
+      return_tensors='pt',
+      truncation=True
+  )
+
+  test_seq = torch.tensor(tokens_test['input_ids'])
+  test_mask = torch.tensor(tokens_test['attention_mask'])
+
+  # get predictions for test data
+  with torch.no_grad():
+    outputs = model(test_seq.cuda(), test_mask.cuda())
+    outputs = outputs.logits.detach().cpu().numpy()
+
+  softmax_score = softmax(outputs)
+  # pred_class_dict = {k:v for k, v in zip(label2idx.keys(), softmax_score[0])}
+  softmax_score = np.argmax(softmax_score, axis=2)[0]
+
+inputs = tokenizer(tobe_splitted, return_tensors="pt", padding=True).to(device)
+translated = model.generate(**inputs, max_length=128)
+tran = [tokenizer.decode(t, skip_special_tokens=True) for t in translated]
+for c, t in zip(tobe_translated, tran):
+    print(f'{c}: {t}')
+```
+講筵官: Lecturer<br>
+判司簿尉: Supervisor of the Commandant of Records<br>
+散騎常侍: Policy Advisor<br>
+殿中省尚輦奉御: Chief Steward of the Palace Administration<br>
+
+work-in-progress
+
+
+### <font color="IndianRed">Authors </font>
+Queenie Luo (queenieluo[at]g.harvard.edu)
+<br>
+Hongsu Wang
+<br>
+Peter Bol
+<br>
+CBDB Group
+
+### <font color="IndianRed">License </font>
+Copyright (c) 2023 CBDB
+
+Except where otherwise noted, content on this repository is licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0).
+To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-sa/4.0/ or
+send a letter to Creative Commons, PO Box 1866, Mountain View, CA 94042, USA.