Creating ReadMe.md
Browse files
README.md
CHANGED
@@ -1,3 +1,71 @@
|
|
1 |
---
|
2 |
license: llama2
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: llama2
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
library_name: transformers
|
6 |
+
pipeline_tag: text2text-generation
|
7 |
+
tags:
|
8 |
+
- CodeMate
|
9 |
+
- Code
|
10 |
---
|
11 |
+
|
12 |
+
# **CodeMate-v0.1**
|
13 |
+
|
14 |
+
CodeMate-v0.1 is an intelligent programming assistant developed by [CodeMate](https://codemate.ai).
|
15 |
+
This model aims to assist users in generating high-quality code solutions for programming problems.
|
16 |
+
Please note that this model is currently in version 0.1.
|
17 |
+
|
18 |
+
## Model Details
|
19 |
+
|
20 |
+
- **Training Data:** Exclusively fine-tuned on a proprietary dataset of 1.8 billion tokens of high-quality programming problems and solutions.
|
21 |
+
- The dataset was generated manually and is internal to CodeMate.
|
22 |
+
|
23 |
+
- **Training Techniques:** The model was fine-tuned using Flash Attention 2, trained over 15 hours on 40 A100-80GB GPUs.
|
24 |
+
- A sequence length of 8096 tokens was used during training.
|
25 |
+
|
26 |
+
- **Multilingual Support:** CodeMate-v0.1 is proficient in multiple programming languages, including Python, C/C++, TypeScript, Java, and more.
|
27 |
+
|
28 |
+
## How to Get Started with the Model
|
29 |
+
|
30 |
+
Make sure to install Transformers from the main git branch:
|
31 |
+
|
32 |
+
```bash
|
33 |
+
pip install git+https://github.com/huggingface/transformers.git
|
34 |
+
```
|
35 |
+
|
36 |
+
## How to Prompt the Model
|
37 |
+
|
38 |
+
This model accepts prompts in the Alpaca/Vicuna instruction format. For example:
|
39 |
+
|
40 |
+
```markdown
|
41 |
+
### System Prompt
|
42 |
+
You are an intelligent programming assistant.
|
43 |
+
|
44 |
+
### User Message
|
45 |
+
Implement a linked list in C++
|
46 |
+
|
47 |
+
### Assistant
|
48 |
+
...
|
49 |
+
```
|
50 |
+
|
51 |
+
## Load the Model:
|
52 |
+
|
53 |
+
To load the model, utilize the following Python script:
|
54 |
+
|
55 |
+
```python
|
56 |
+
from transformers import AutoTokenizer, AutoModelForCausalLM
|
57 |
+
|
58 |
+
# Initialize the model
|
59 |
+
model_path = "codemateai/CodeMate-v0.1"
|
60 |
+
model = AutoModelForCausalLM.from_pretrained(model_path, device_map="auto")
|
61 |
+
tokenizer = AutoTokenizer.from_pretrained(model_path)
|
62 |
+
|
63 |
+
# ... generate response ...
|
64 |
+
|
65 |
+
```
|
66 |
+
|
67 |
+
## Bias, Risks, and Limitations
|
68 |
+
|
69 |
+
This model has undergone very limited testing. CodeMate recommends additional safety testing before any real-world deployments.
|
70 |
+
|
71 |
+
For more information and updates, visit the [CodeMate website](https://codemate.ai).
|