Update README.md
Browse files
README.md
CHANGED
|
@@ -11,7 +11,6 @@ Generating higher quality variable names for code by renaming masked variable na
|
|
| 11 |
|
| 12 |
|
| 13 |
- **Developed by:** SMART Lab, Dalhousie University
|
| 14 |
-
- **Funded by:** [More Information Needed]
|
| 15 |
- **Model type:** Masked Language model
|
| 16 |
- **Language(s) (NLP):** Coded in Python to handle Java code
|
| 17 |
- **Finetuned from model:** GraphCodeBERT
|
|
@@ -35,21 +34,20 @@ The code snippets must ideally be entire classes for best results. A prediction
|
|
| 35 |
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 36 |
This non-fine-tuned version of the model is designed for generic code completion tasks. The fine-tuned model is designed to focus solely on identifier names.<br>
|
| 37 |
Ensure all instances of a particular variable name are masked.
|
| 38 |
-
[More Information Needed]
|
| 39 |
|
| 40 |
## Bias, Risks, and Limitations
|
| 41 |
|
| 42 |
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 43 |
Training is only done for a relatively small dataset and few epochs, and thus, the model might be under-trained. <br>
|
| 44 |
Even with the correct output, the syntax of the model can be occasionally dubious.<br>
|
| 45 |
-
|
| 46 |
|
| 47 |
|
| 48 |
### Recommendations
|
| 49 |
|
| 50 |
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 51 |
|
| 52 |
-
Use model as described and verify outputs before using them.
|
| 53 |
## How to Get Started with the Model
|
| 54 |
|
| 55 |
Clone the repository and load model state dict using 'model_26_2'
|
|
|
|
| 11 |
|
| 12 |
|
| 13 |
- **Developed by:** SMART Lab, Dalhousie University
|
|
|
|
| 14 |
- **Model type:** Masked Language model
|
| 15 |
- **Language(s) (NLP):** Coded in Python to handle Java code
|
| 16 |
- **Finetuned from model:** GraphCodeBERT
|
|
|
|
| 34 |
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
| 35 |
This non-fine-tuned version of the model is designed for generic code completion tasks. The fine-tuned model is designed to focus solely on identifier names.<br>
|
| 36 |
Ensure all instances of a particular variable name are masked.
|
|
|
|
| 37 |
|
| 38 |
## Bias, Risks, and Limitations
|
| 39 |
|
| 40 |
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
| 41 |
Training is only done for a relatively small dataset and few epochs, and thus, the model might be under-trained. <br>
|
| 42 |
Even with the correct output, the syntax of the model can be occasionally dubious.<br>
|
| 43 |
+
The model is not perfect, and identifier renamings must be reviewed till performance in test settings is not evaluated.
|
| 44 |
|
| 45 |
|
| 46 |
### Recommendations
|
| 47 |
|
| 48 |
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
| 49 |
|
| 50 |
+
Use the model as described and verify outputs before using them.
|
| 51 |
## How to Get Started with the Model
|
| 52 |
|
| 53 |
Clone the repository and load model state dict using 'model_26_2'
|