Update README.md

README.md

@@ -25,153 +25,195 @@ UniXcoder is a unified cross-modal pre-trained model that leverages multimodal d
  
 # Uses
  
- 
- 
-## Direct Use
- Feature Engineering
- 
- 
-## Downstream Use [Optional]
- 
-More information needed
- 
-## Out-of-Scope Use
- 
-More information needed
- 
-# Bias, Risks, and Limitations
- 
-Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
- 
- 
-## Recommendations
- 
- 
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
- 
- 
-# Training Details
- 
-## Training Data
- 
-More information needed
- 
-## Training Procedure
- 
-### Preprocessing
- 
-More information needed
- 
-### Speeds, Sizes, Times
-More information needed
- 
-# Evaluation
- 
- 
-## Testing Data, Factors & Metrics
- 
-### Testing Data
-More information needed
- 
-### Factors
- 
-The model creators note in the [associated paper](https://arxiv.org/abs/2203.03850):
- 
-> UniXcoder has slightly worse BLEU-4 scores on both code summarization and generation tasks. The main reasons may come from two aspects. One is the amount of NL-PL pairs in the pre-training data
- 
-### Metrics
- 
-The model creators note in the [associated paper](https://arxiv.org/abs/2203.03850):
- 
-> We evaluate UniXcoder on five tasks over nine public datasets, including two understanding tasks, two generation tasks and an autoregressive task. To further evaluate the performance of code fragment embeddings, we also propose a new task called zero-shot code-to-code search.
- 
-## Results 
- 
-The model creators note in the [associated paper](https://arxiv.org/abs/2203.03850):
- 
->Taking zero-shot code-code search task as an example, after removing contrastive learning, the performance drops from 20.45% to 13.73%.
- 
-# Model Examination
- 
-More information needed
- 
-# Environmental Impact
- 
- 
- 
-Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
- 
-- **Hardware Type:** More information needed
-- **Hours used:** More information needed
-- **Cloud Provider:** More information needed
-- **Compute Region:** More information needed
-- **Carbon Emitted:** More information needed
- 
-# Technical Specifications [optional]
- 
-## Model Architecture and Objective
- 
-More information needed
- 
-## Compute Infrastructure
- 
-More information needed
- 
-### Hardware
- 
-More information needed
- 
-### Software
- 
-More information needed
- 
-# Citation
- 
- 
-**BibTeX:**
- ```
-@misc{https://doi.org/10.48550/arxiv.2203.03850,
-  doi = {10.48550/ARXIV.2203.03850},
-  
-  url = {https://arxiv.org/abs/2203.03850},
-  
-  author = {Guo, Daya and Lu, Shuai and Duan, Nan and Wang, Yanlin and Zhou, Ming and Yin, Jian},
-  
-  keywords = {Computation and Language (cs.CL), Programming Languages (cs.PL), Software Engineering (cs.SE), FOS: Computer and information sciences, FOS: Computer and information sciences},
-  
-  title = {UniXcoder: Unified Cross-Modal Pre-training for Code 
+## 1. Dependency
+
+- pip install torch
+- pip install transformers
+
+## 2. Quick Tour
+We implement a class to use UniXcoder and you can follow the code to build UniXcoder.
+You can download the class by
+```shell
+wget https://raw.githubusercontent.com/microsoft/CodeBERT/master/UniXcoder/unixcoder.py
 ```
- 
- 
-# Glossary [optional]
- 
-More information needed
- 
-# More Information [optional]
- 
-More information needed
- 
-# Model Card Authors [optional]
- 
-Microsoft Team in collaboration with Ezi Ozoani and the Hugging Face Team.
- 
-# Model Card Contact
- 
-More information needed
- 
-# How to Get Started with the Model
- 
-Use the code below to get started with the model.
- 
-<details>
-<summary> Click to expand </summary>
 
 ```python
-from transformers import AutoTokenizer, AutoModel
- 
-tokenizer = AutoTokenizer.from_pretrained("microsoft/unixcoder-base")
- 
-model = AutoModel.from_pretrained("microsoft/unixcoder-base")
- 
- ```
-</details>
+import torch
+from unixcoder import UniXcoder
+
+device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
+model = UniXcoder("microsoft/unixcoder-base")
+model.to(device)
+```
+
+In the following, we will give zero-shot examples for several tasks under different mode, including **code search (encoder-only)**, **code completion (decoder-only)**, **function name prediction (encoder-decoder)** , **API recommendation (encoder-decoder)**, **code summarization (encoder-decoder)**.
+
+## 3. Encoder-only Mode
+
+For encoder-only mode, we give an example of **code search**.
+
+### 1) Code and NL Embeddings
+
+Here, we give an example to obtain code fragment embedding from CodeBERT.
+
+```python
+# Encode maximum function
+func = "def f(a,b): if a>b: return a else return b"
+tokens_ids = model.tokenize([func],max_length=512,mode="<encoder-only>")
+source_ids = torch.tensor(tokens_ids).to(device)
+tokens_embeddings,max_func_embedding = model(source_ids)
+
+# Encode minimum function
+func = "def f(a,b): if a<b: return a else return b"
+tokens_ids = model.tokenize([func],max_length=512,mode="<encoder-only>")
+source_ids = torch.tensor(tokens_ids).to(device)
+tokens_embeddings,min_func_embedding = model(source_ids)
+
+# Encode NL
+nl = "return maximum value"
+tokens_ids = model.tokenize([nl],max_length=512,mode="<encoder-only>")
+source_ids = torch.tensor(tokens_ids).to(device)
+tokens_embeddings,nl_embedding = model(source_ids)
+
+print(max_func_embedding.shape)
+print(max_func_embedding)
+```
+
+```python
+torch.Size([1, 768])
+tensor([[ 8.6533e-01, -1.9796e+00, -8.6849e-01,  4.2652e-01, -5.3696e-01,
+         -1.5521e-01,  5.3770e-01,  3.4199e-01,  3.6305e-01, -3.9391e-01,
+         -1.1816e+00,  2.6010e+00, -7.7133e-01,  1.8441e+00,  2.3645e+00,
+				 ...,
+         -2.9188e+00,  1.2555e+00, -1.9953e+00, -1.9795e+00,  1.7279e+00,
+          6.4590e-01, -5.2769e-02,  2.4965e-01,  2.3962e-02,  5.9996e-02,
+          2.5659e+00,  3.6533e+00,  2.0301e+00]], device='cuda:0',
+       grad_fn=<DivBackward0>)
+```
+
+### 2) Similarity between code and NL
+
+Now, we calculate cosine similarity between NL and two functions. Although the difference of two functions is only a operator (```<``` and ```>```), UniXcoder can distinguish them.
+
+```python
+# Normalize embedding
+norm_max_func_embedding = torch.nn.functional.normalize(max_func_embedding, p=2, dim=1)
+norm_min_func_embedding = torch.nn.functional.normalize(min_func_embedding, p=2, dim=1)
+norm_nl_embedding = torch.nn.functional.normalize(nl_embedding, p=2, dim=1)
+
+max_func_nl_similarity = torch.einsum("ac,bc->ab",norm_max_func_embedding,norm_nl_embedding)
+min_func_nl_similarity = torch.einsum("ac,bc->ab",norm_min_func_embedding,norm_nl_embedding)
+
+print(max_func_nl_similarity)
+print(min_func_nl_similarity)
+```
+
+```python
+tensor([[0.3002]], device='cuda:0', grad_fn=<ViewBackward>)
+tensor([[0.1881]], device='cuda:0', grad_fn=<ViewBackward>)
+```
+
+## 3. Decoder-only Mode
+
+For decoder-only mode, we give an example of **code completion**.
+
+```python
+context = """
+def f(data,file_path):
+    # write json data into file_path in python language
+"""
+tokens_ids = model.tokenize([context],max_length=512,mode="<decoder-only>")
+source_ids = torch.tensor(tokens_ids).to(device)
+prediction_ids = model.generate(source_ids, decoder_only=True, beam_size=3, max_length=128)
+predictions = model.decode(prediction_ids)
+print(context+predictions[0][0])
+```
+
+```python
+def f(data,file_path):
+    # write json data into file_path in python language
+    data = json.dumps(data)
+    with open(file_path, 'w') as f:
+        f.write(data)
+```
+
+## 4. Encoder-Decoder Mode
+
+For encoder-decoder mode, we give two examples including: **function name prediction**, **API recommendation**, **code summarization**.
+
+### 1) **Function Name Prediction**
+
+```python
+context = """
+def <mask0>(data,file_path):
+    data = json.dumps(data)
+    with open(file_path, 'w') as f:
+        f.write(data)
+"""
+tokens_ids = model.tokenize([context],max_length=512,mode="<encoder-decoder>")
+source_ids = torch.tensor(tokens_ids).to(device)
+prediction_ids = model.generate(source_ids, decoder_only=False, beam_size=3, max_length=128)
+predictions = model.decode(prediction_ids)
+print([x.replace("<mask0>","").strip() for x in predictions[0]])
+```
+
+```python
+['write_json', 'write_file', 'to_json']
+```
+
+### 2) API Recommendation
+
+```python
+context = """
+def write_json(data,file_path):
+    data = <mask0>(data)
+    with open(file_path, 'w') as f:
+        f.write(data)
+"""
+tokens_ids = model.tokenize([context],max_length=512,mode="<encoder-decoder>")
+source_ids = torch.tensor(tokens_ids).to(device)
+prediction_ids = model.generate(source_ids, decoder_only=False, beam_size=3, max_length=128)
+predictions = model.decode(prediction_ids)
+print([x.replace("<mask0>","").strip() for x in predictions[0]])
+```
+
+```python
+['json.dumps', 'json.loads', 'str']
+```
+
+### 3) Code Summarization
+
+```python
+context = """
+# <mask0>
+def write_json(data,file_path):
+    data = json.dumps(data)
+    with open(file_path, 'w') as f:
+        f.write(data)
+"""
+tokens_ids = model.tokenize([context],max_length=512,mode="<encoder-decoder>")
+source_ids = torch.tensor(tokens_ids).to(device)
+prediction_ids = model.generate(source_ids, decoder_only=False, beam_size=3, max_length=128)
+predictions = model.decode(prediction_ids)
+print([x.replace("<mask0>","").strip() for x in predictions[0]])
+```
+
+```python
+['Write JSON to file', 'Write json to file', 'Write a json file']
+```
+
+
+
+
+# Reference
+If you use this code or UniXcoder, please consider citing us.
+
+<pre><code>@article{guo2022unixcoder,
+  title={UniXcoder: Unified Cross-Modal Pre-training for Code Representation},
+  author={Guo, Daya and Lu, Shuai and Duan, Nan and Wang, Yanlin and Zhou, Ming and Yin, Jian},
+  journal={arXiv preprint arXiv:2203.03850},
+  year={2022}
+}</code></pre>
+
+
+