Update README.md
Browse files
README.md
CHANGED
@@ -9,6 +9,10 @@ tags:
|
|
9 |
- gguf
|
10 |
- f32
|
11 |
- f16
|
|
|
|
|
|
|
|
|
12 |
- humaneval
|
13 |
license: mit
|
14 |
---
|
@@ -17,14 +21,16 @@ license: mit
|
|
17 |
|
18 |
WaveCoder 🌊 is a series of large language models (LLMs) for the coding domain.
|
19 |
|
20 |
-
Apologies for the incomplete model details, the GitHub repo doesn't exist and I'm currently trying to quant all the models.
|
21 |
-
|
22 |
## Model Details
|
23 |
|
24 |
### Model Description
|
25 |
|
26 |
WaveCoder 🌊 is a series of large language models (LLMs) for the coding domain, designed to solve relevant problems in the field of code through instruction-following learning. Its training dataset was generated from a subset of code-search-net data using a generator-discriminator framework based on LLMs that we proposed, covering four general code-related tasks: code generation, code summary, code translation, and code repair.
|
27 |
|
|
|
|
|
|
|
|
|
28 |
- **Developed by:** Yu, Zhaojian and Zhang, Xin and Shang, Ning and Huang, Yangyu and Xu, Can and Zhao, Yishujie and Hu, Wenxiang and Yin, Qiufeng
|
29 |
- **Model type:** Large Language Model
|
30 |
- **Language(s) (NLP):** English
|
@@ -32,12 +38,12 @@ WaveCoder 🌊 is a series of large language models (LLMs) for the coding domain
|
|
32 |
|
33 |
### Model Sources
|
34 |
|
35 |
-
- **Repository:** [
|
36 |
-
- **Paper
|
37 |
-
- **Demo
|
38 |
|
39 |
## Uses
|
40 |
|
41 |
-
Coding
|
42 |
|
43 |
## Original: [https://huggingface.co/microsoft/wavecoder-ds-6.7b](https://huggingface.co/microsoft/wavecoder-ds-6.7b)
|
|
|
9 |
- gguf
|
10 |
- f32
|
11 |
- f16
|
12 |
+
- q2
|
13 |
+
- q8
|
14 |
+
- q6
|
15 |
+
- q4_k_m
|
16 |
- humaneval
|
17 |
license: mit
|
18 |
---
|
|
|
21 |
|
22 |
WaveCoder 🌊 is a series of large language models (LLMs) for the coding domain.
|
23 |
|
|
|
|
|
24 |
## Model Details
|
25 |
|
26 |
### Model Description
|
27 |
|
28 |
WaveCoder 🌊 is a series of large language models (LLMs) for the coding domain, designed to solve relevant problems in the field of code through instruction-following learning. Its training dataset was generated from a subset of code-search-net data using a generator-discriminator framework based on LLMs that we proposed, covering four general code-related tasks: code generation, code summary, code translation, and code repair.
|
29 |
|
30 |
+
WaveCoder-ds = Trained using CodeOcean dataset
|
31 |
+
WaveCoder-pro = Trained using GPT-4 synthetic data
|
32 |
+
WaveCoder-ultra = Trained using enhanced GPT-4 synthetic data
|
33 |
+
|
34 |
- **Developed by:** Yu, Zhaojian and Zhang, Xin and Shang, Ning and Huang, Yangyu and Xu, Can and Zhao, Yishujie and Hu, Wenxiang and Yin, Qiufeng
|
35 |
- **Model type:** Large Language Model
|
36 |
- **Language(s) (NLP):** English
|
|
|
38 |
|
39 |
### Model Sources
|
40 |
|
41 |
+
- **Repository:** [https://huggingface.co/microsoft/wavecoder-ds-6.7b](https://huggingface.co/microsoft/wavecoder-ds-6.7b)
|
42 |
+
- **Paper :** [More Information Needed]
|
43 |
+
- **Demo :** [More Information Needed]
|
44 |
|
45 |
## Uses
|
46 |
|
47 |
+
Coding/Refactoring/Cleanup/Fixing Code
|
48 |
|
49 |
## Original: [https://huggingface.co/microsoft/wavecoder-ds-6.7b](https://huggingface.co/microsoft/wavecoder-ds-6.7b)
|