mchochlov/codebert-base-cd-ft

1---2pipeline_tag: sentence-similarity3tags:4- sentence-transformers5- feature-extraction6- sentence-similarity7- transformers8---9 10# mchochlov/codebert-base-cd-ft11 12This is a [sentence-transformers](https://www.SBERT.net) model: It maps code to a 768 dimensional dense vector space and is specifically fine tuned towards clone detection using contrastive learning on parts of BigCloneBench code.13 14<!--- Describe your model here -->15 16## Usage (Sentence-Transformers)17 18Using this model becomes easy when you have [sentence-transformers](https://www.SBERT.net) installed:19 20```21pip install -U sentence-transformers22```23 24Then you can use the model like this:25 26```python27from sentence_transformers import SentenceTransformer28code_fragments = [...]29 30model = SentenceTransformer('mchochlov/codebert-base-cd-ft')31embeddings = model.encode(code_fragments)32print(embeddings)33```34 35 36 37## Usage (HuggingFace Transformers)38Without [sentence-transformers](https://www.SBERT.net), you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.39 40```python41from transformers import AutoTokenizer, AutoModel42import torch43 44 45#Mean Pooling - Take attention mask into account for correct averaging46def mean_pooling(model_output, attention_mask):47    token_embeddings = model_output[0] #First element of model_output contains all token embeddings48    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()49    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)50 51 52# Sentences we want sentence embeddings for53sentences = ['This is an example sentence', 'Each sentence is converted']54 55# Load model from HuggingFace Hub56tokenizer = AutoTokenizer.from_pretrained('mchochlov/codebert-base-cd-ft')57model = AutoModel.from_pretrained('mchochlov/codebert-base-cd-ft')58 59# Tokenize sentences60encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')61 62# Compute token embeddings63with torch.no_grad():64    model_output = model(**encoded_input)65 66# Perform pooling. In this case, max pooling.67sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])68 69print("Sentence embeddings:")70print(sentence_embeddings)71```72 73 74 75## Evaluation Results76 77<!--- Describe how your model was evaluated -->78 79For an automated evaluation of this model, see the *Sentence Embeddings Benchmark*: [https://seb.sbert.net](https://seb.sbert.net?model_name=mchochlov/codebert-base-cd-ft)80 81 82 83## Full Model Architecture84```85SentenceTransformer(86  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: RobertaModel 87  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})88)89```90 91## Citing & Authors92 93<!--- Describe where people can find more information -->94Please cite this paper if using the model.95```latex96@inproceedings{chochlov2022using,97  title={Using a Nearest-Neighbour, BERT-Based Approach for Scalable Clone Detection},98  author={Chochlov, Muslim and Ahmed, Gul Aftab and Patten, James Vincent and Lu, Guoxian and Hou, Wei and Gregg, David and Buckley, Jim},99  booktitle={2022 IEEE International Conference on Software Maintenance and Evolution (ICSME)},100  pages={582--591},101  year={2022},102  organization={IEEE}103}104```