Update README.md
Browse files
README.md
CHANGED
@@ -11,12 +11,62 @@ language:
|
|
11 |
- en
|
12 |
---
|
13 |
|
14 |
-
#
|
15 |
|
16 |
-
|
17 |
-
- **License:** apache-2.0
|
18 |
-
- **Finetuned from model :** unsloth/deepseek-r1-distill-qwen-7b-unsloth-bnb-4bit
|
19 |
|
20 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
21 |
|
22 |
-
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
|
|
11 |
- en
|
12 |
---
|
13 |
|
14 |
+
# MaWared HR Reasoning Model
|
15 |
|
16 |
+
## Model Details
|
|
|
|
|
17 |
|
18 |
+
- **Base Model:** [unsloth/deepseek-r1-distill-qwen-7b-unsloth-bnb-4bit](https://huggingface.co/unsloth/deepseek-r1-distill-qwen-7b-unsloth-bnb-4bit)
|
19 |
+
- **Finetuned by:** Daemontatox
|
20 |
+
- **License:** Apache-2.0
|
21 |
+
- **Language:** English
|
22 |
+
- **Tags:** text-generation-inference, transformers, unsloth, qwen2, trl
|
23 |
+
|
24 |
+
## Overview
|
25 |
+
|
26 |
+
This model is a finetuned version of the `deepseek-r1-distill-qwen-7b` model, optimized for MaWared HR reasoning. It was trained using [Unsloth](https://github.com/unslothai/unsloth) and Hugging Face's TRL library, enabling 2x faster training performance.
|
27 |
+
|
28 |
+
## Features
|
29 |
+
|
30 |
+
- **HR Query Reasoning:** Provides logical and well-structured responses to complex HR-related inquiries.
|
31 |
+
- **Decision Support:** Assists HR professionals in making informed decisions based on policies and regulations.
|
32 |
+
- **Enhanced Performance:** Optimized for deep reasoning and contextual understanding in HR-related scenarios.
|
33 |
+
|
34 |
+
## Installation
|
35 |
+
|
36 |
+
To use this model, install the required dependencies:
|
37 |
+
|
38 |
+
```bash
|
39 |
+
pip install torch transformers accelerate unsloth
|
40 |
+
|
41 |
+
```
|
42 |
+
|
43 |
+
## Usage
|
44 |
+
You can load and use the model with the following Python snippet:
|
45 |
+
```
|
46 |
+
from transformers import AutoModelForCausalLM, AutoTokenizer
|
47 |
+
import torch
|
48 |
+
|
49 |
+
model_name = "Daemontatox/mawared-hr-reasoning"
|
50 |
+
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
51 |
+
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
|
52 |
+
|
53 |
+
input_text = "How should I handle a conflict between employees?"
|
54 |
+
inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
|
55 |
+
output = model.generate(**inputs, max_length=100)
|
56 |
+
response = tokenizer.decode(output[0], skip_special_tokens=True)
|
57 |
+
print(response)
|
58 |
+
```
|
59 |
+
|
60 |
+
|
61 |
+
## Acknowledgments
|
62 |
+
This model was developed using Unsloth and Hugging Face's TRL library. Special thanks to the open-source community for their contributions.
|
63 |
+
|
64 |
+
License
|
65 |
+
This model is licensed under the Apache-2.0 license.
|
66 |
+
|
67 |
+
vbnet
|
68 |
+
```
|
69 |
+
|
70 |
+
Let me know if you need any modifications! 🚀
|
71 |
+
```
|
72 |
|
|