hungnm
/

viBioGPT-7B-instruct-qlora-adapter

@@ -4,206 +4,127 @@ base_model: Viet-Mistral/Vistral-7B-Chat
 language:
 - vi
 tags:
-- medical
-- 'llm '
 ---
-# Model Card for Model ID
-<!-- Provide a quick summary of what the model is/does. -->
-## Model Details
-### Model Description
-<!-- Provide a longer summary of what this model is. -->
-- **Developed by:** [More Information Needed]
-- **Funded by [optional]:** [More Information Needed]
-- **Shared by [optional]:** [More Information Needed]
-- **Model type:** [More Information Needed]
-- **Language(s) (NLP):** [More Information Needed]
-- **License:** [More Information Needed]
-- **Finetuned from model [optional]:** [More Information Needed]
-### Model Sources [optional]
-<!-- Provide the basic links for the model. -->
-- **Repository:** [More Information Needed]
-- **Paper [optional]:** [More Information Needed]
-- **Demo [optional]:** [More Information Needed]
-## Uses
-<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
-### Direct Use
-<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
-[More Information Needed]
-### Downstream Use [optional]
-<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
-[More Information Needed]
-### Out-of-Scope Use
-<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
-[More Information Needed]
-## Bias, Risks, and Limitations
-<!-- This section is meant to convey both technical and sociotechnical limitations. -->
-[More Information Needed]
-### Recommendations
-<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
-Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
-## How to Get Started with the Model
-Use the code below to get started with the model.
-[More Information Needed]
-## Training Details
-### Training Data
-<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
-[More Information Needed]
-### Training Procedure
-<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
-#### Preprocessing [optional]
-[More Information Needed]
-#### Training Hyperparameters
-- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
-#### Speeds, Sizes, Times [optional]
-<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
-[More Information Needed]
-## Evaluation
-<!-- This section describes the evaluation protocols and provides the results. -->
-### Testing Data, Factors & Metrics
-#### Testing Data
-<!-- This should link to a Dataset Card if possible. -->
-[More Information Needed]
-#### Factors
-<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
-[More Information Needed]
-#### Metrics
-<!-- These are the evaluation metrics being used, ideally with a description of why. -->
-[More Information Needed]
-### Results
-[More Information Needed]
-#### Summary
-## Model Examination [optional]
-<!-- Relevant interpretability work for the model goes here -->
-[More Information Needed]
-## Environmental Impact
-<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
-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 [optional]
-<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
-**BibTeX:**
-[More Information Needed]
-**APA:**
-[More Information Needed]
-## Glossary [optional]
-<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
-[More Information Needed]
-## More Information [optional]
-[More Information Needed]
-## Model Card Authors [optional]
-[More Information Needed]
-## Model Card Contact
-[More Information Needed]
 ### Framework versions
-- PEFT 0.7.1

 language:
 - vi
 tags:
+- LLMs
+- Vietnamese
+- Medical
+- Health
+- Vistral
+- NLP
+license: apache-2.0
+datasets:
+- hungnm/vietnamese-medical-qa
 ---
+## Model Description
+**viBioGPT-7B-instruct** is a Vietnamese Large Language Model (LLM) fine-tuned for the task of Question Answering within
+the medical and healthcare domain. This model uses pre-trained [Vistral-Chat-7B](https://huggingface.co/Viet-Mistral/Vistral-7B-Chat), then QLora technique
+to fine-tune.
+training dataset:[hungnm/vietnamese-medical-qa](https://huggingface.co/datasets/hungnm/vietnamese-medical-qa)
+## How to Use
+Install libraries
+```shell
+pip install peft==0.7.1 bitsandbytes==0.41.3.post2 transformers==4.36.2 torch==2.1.2
+```
+Because this adapter uses pretrained [Viet-Mistral/Vistral-7B-Chat](https://huggingface.co/Viet-Mistral/Vistral-7B-Chat), ensure that you granted access to that model and set your huggingface token in code.
+```python
+import torch
+from peft import PeftModel
+from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
+HF_TOKEN = "<your_hf_token>"
+model_name = "Viet-Mistral/Vistral-7B-Chat"
+adapter = "hungnm/viBioGPT-7B-instruct-qlora-adapter"
+compute_dtype = getattr(torch, "bfloat16")
+bnb_config = BitsAndBytesConfig(
+    load_in_4bit=True,
+    bnb_4bit_quant_type="nf4",
+    bnb_4bit_compute_dtype=compute_dtype,
+    bnb_4bit_use_double_quant=True,
+)
+model = AutoModelForCausalLM.from_pretrained(model_name,
+                                             quantization_config=bnb_config,
+                                             device_map={"": 0},
+                                             token=HF_TOKEN
+                                             )
+model = PeftModel.from_pretrained(model, adapter)
+# load and config tokenizer
+tokenizer = AutoTokenizer.from_pretrained(model_name,
+                                          token=HF_TOKEN)
+tokenizer.padding_side = "left"
+tokenizer.pad_token_id = tokenizer.eos_token_id
+system_prompt = ("Bạn là một trợ lý ảo AI trong lĩnh vực Y học, Sức Khỏe. Tên của bạn là AI-Doctor. "
+                 "Nhiệm vụ của bạn là trả lời các thắc mắc hoặc các câu hỏi về Y học, Sức khỏe.")
+question = "tôi có một ít nhân sâm nhưng đang bị viêm dạ dày. Vậy tôi có nên ăn nhân sâm ko?"
+conversation = [
+    {
+        "role": "system",
+        "content": system_prompt},
+    {
+        "role": "user",
+        "content": question
+    }]
+instruction_str = tokenizer.apply_chat_template(conversation=conversation,
+                                                tokenize=False)
+token_ids = tokenizer([instruction_str], return_tensors="pt")["input_ids"]
+token_ids = token_ids.to(model.device)
+outputs = model.generate(input_ids=token_ids,
+                         max_new_tokens=768,
+                         do_sample=True,
+                         temperature=0.1,
+                         top_p=0.95,
+                         top_k=40,
+                         repetition_penalty=1.2)
+all_token_ids = outputs[0].tolist()
+output_token_ids = all_token_ids[token_ids.shape[-1]:]
+output = tokenizer.decode(output_token_ids)
+print(output)
+```
+```text
+Chào anh!
+Nhân sâm được biết đến như loại thảo dược quý hiếm và rất tốt cho sức khoẻ con người tuy nhiên không phải ai cũng dùng được nó đặc biệt với những bệnh nhân đau dạ dày thì càng cần thận trọng khi sử dụng vì nếu lạm dụng sẽ gây ra nhiều tác hại nghiêm trọng tới hệ tiêu hoá nói chung và tình trạng đau dạ dày nói riêng .
+Vì vậy trước tiên anh hãy điều trị dứt điểm căn bênh này rồi mới nghĩ tới việc bổ sung thêm dinh dưỡng từ nhân sâm nhé !
+Chúc anh mau khỏi bệnh ạ!
+```
+### Disclaimer
+Despite thorough testing, our model may still carry risks such as hallucination, toxic content, and bias. We urge users to recognize and address these risks before use. Users are responsible for compliance with regulations, and the authors disclaim liability for any resulting damages.**
 ### Framework versions
+- accelerate==0.21.0
+- sentencepiece==0.1.99
+- transformers==4.36.2
+- peft==0.7.1
+- bitsandbytes==0.41.3.post2
+- wandb==0.16.1
+- numpy==1.26.2
+- datasets==2.15.0
+- python-dotenv==1.0.1
+- flash-attn==2.5.3
+## Citation
+If you find our project helpful, please star our repo and cite our work. Thanks!
+```bibtex
+@misc{viBioGPT,
+      title={viBioGPT: A Vietnamese Large Language Model for Biomedical Question Answering},
+      author={Hung Nguyen},
+      year={2024},
+}```