Update README.md

README.md

@@ -7,50 +7,6 @@ language:
 - ja
 ---
 
-SPLADE-japanese-v2 !!
-
-Difference between splade-japanese v1 and v2
-- initialize [tohoku-nlp/bert-base-japanese-v3](https://huggingface.co/tohoku-nlp/bert-base-japanese-v3)
-- knowledge distillation from cross-encoder
-- [mMARCO](https://github.com/unicamp-dl/mMARCO) Japanese dataset and use [bclavie/mmarco-japanese-hard-negatives](https://huggingface.co/datasets/bclavie/mmarco-japanese-hard-negatives) as hard negatives
-
-
-you need to install 
-'''
-!pip install fugashi ipadic unidic-lite
-'''
-
-```python
-from transformers import AutoModelForMaskedLM,AutoTokenizer
-import torch
-import numpy as np
-
-model = AutoModelForMaskedLM.from_pretrained("aken12/splade-japanesev2-epoch5") 
-tokenizer = AutoTokenizer.from_pretrained("aken12/splade-japanesev2-epoch5")
-vocab_dict = {v: k for k, v in tokenizer.get_vocab().items()}
-
-def encode_query(query):
-    query = tokenizer(query, return_tensors="pt")
-    output = model(**query, return_dict=True).logits
-    output, _ = torch.max(torch.log(1 + torch.relu(output)) * query['attention_mask'].unsqueeze(-1), dim=1)
-    return output
-
-with torch.no_grad():
-    model_output = encode_query(query="筑波大学では何の研究が行われているか？")
-
-reps = model_output
-idx = torch.nonzero(reps[0], as_tuple=False)
-
-dict_splade = {}
-for i in idx:
-    token_value = reps[0][i[0]].item()
-    if token_value > 0:
-        token = vocab_dict[int(i[0])]
-        dict_splade[token] = float(token_value)
-
-sorted_dict_splade = sorted(dict_splade.items(), key=lambda item: item[1], reverse=True)
-for token, value in sorted_dict_splade:
-    print(token, value)
 
 ```