README.md

---
language:
- pt
license: apache-2.0
library_name: transformers
tags:
- Misral
- Portuguese
- 7b
- chat
- portugues
base_model: mistralai/Mistral-7B-Instruct-v0.2
datasets:
- rhaymison/ultrachat-easy-use
pipeline_tag: text-generation
model-index:
- name: Mistral-portuguese-luana-7b-chat
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: ENEM Challenge (No Images)
      type: eduagarcia/enem_challenge
      split: train
      args:
        num_few_shot: 3
    metrics:
    - type: acc
      value: 59.13
      name: accuracy
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: BLUEX (No Images)
      type: eduagarcia-temp/BLUEX_without_images
      split: train
      args:
        num_few_shot: 3
    metrics:
    - type: acc
      value: 49.24
      name: accuracy
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: OAB Exams
      type: eduagarcia/oab_exams
      split: train
      args:
        num_few_shot: 3
    metrics:
    - type: acc
      value: 36.58
      name: accuracy
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Assin2 RTE
      type: assin2
      split: test
      args:
        num_few_shot: 15
    metrics:
    - type: f1_macro
      value: 90.47
      name: f1-macro
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Assin2 STS
      type: eduagarcia/portuguese_benchmark
      split: test
      args:
        num_few_shot: 15
    metrics:
    - type: pearson
      value: 76.55
      name: pearson
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: FaQuAD NLI
      type: ruanchaves/faquad-nli
      split: test
      args:
        num_few_shot: 15
    metrics:
    - type: f1_macro
      value: 66.75
      name: f1-macro
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HateBR Binary
      type: ruanchaves/hatebr
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: f1_macro
      value: 77.46
      name: f1-macro
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: PT Hate Speech Binary
      type: hate_speech_portuguese
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: f1_macro
      value: 69.45
      name: f1-macro
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: tweetSentBR
      type: eduagarcia-temp/tweetsentbr
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: f1_macro
      value: 59.63
      name: f1-macro
    source:
      url: https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard?query=rhaymison/Mistral-portuguese-luana-7b-chat
      name: Open Portuguese LLM Leaderboard
---

# Mistral-portuguese-luana-7b-chat

<p align="center">
  <img src="https://raw.githubusercontent.com/rhaymisonbetini/huggphotos/main/luana-chat.webp"  width="50%" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
</p>


This model was trained with a superset of 250,000 chat in Portuguese. 
The model comes to help fill the gap in models in Portuguese. Tuned from the Mistral 7b, the model was adjusted mainly for chat.

# How to use

### FULL MODEL : A100
### HALF MODEL: L4
### 8bit or 4bit : T4 or V100

You can use the model in its normal form up to 4-bit quantization. Below we will use both approaches.
Remember that verbs are important in your prompt. Tell your model how to act or behave so that you can guide them along the path of their response. 
Important points like these help models (even smaller models like 7b) to perform much better.

```python
!pip install -q -U transformers
!pip install -q -U accelerate
!pip install -q -U bitsandbytes

from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
model = AutoModelForCausalLM.from_pretrained("rhaymison/Mistral-portuguese-luana-7b-chat", device_map= {"": 0})
tokenizer = AutoTokenizer.from_pretrained("rhaymison/Mistral-portuguese-luana-7b-chat")
model.eval()

```

You can use with Pipeline but in this example i will use such as Streaming
```python

inputs = tokenizer([f"""<s>[INST] Estou pensando em levar minha esposa e meu gatinha para fazer algo diferente. Você teria alguma coisa em mente ?.
[/INST]"""], return_tensors="pt")

inputs.to(model.device)

streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
_ = model.generate(**inputs, streamer=streamer, max_new_tokens=128)

```

If you are having a memory problem such as "CUDA Out of memory", you should use 4-bit or 8-bit quantization.
For the complete model in colab you will need the A100.
If you want to use 4bits or 8bits, T4 or L4 will already solve the problem.

# 4bits example

```python
from transformers import BitsAndBytesConfig
import torch
nb_4bit_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=torch.bfloat16,
    bnb_4bit_use_double_quant=True
)

model = AutoModelForCausalLM.from_pretrained(
    base_model,
    quantization_config=bnb_config,
    device_map={"": 0}
)

```

# LangChain 

<div style="background-color:#f0f0f0; padding:10px; border-radius:10px; font-family: 'Arial';">
  <p><strong><span style="color:#3366cc;">Human:</span></strong> Ola Luana, estou precisando de umas dicas para o final de semana.
  Poderia me ajudar ?</p>

  <p><strong><span style="color:#ff6600;">Luana:</span></strong> Olá! Claro, ficarei feliz em ajudá-lo. O que você tem em mente para o final de semana?</p>

  <p><strong><span style="color:#3366cc;">Human:</span></strong> Estou pensando em levar minha esposa e meu gatinha para fazer algo diferente.
  Você teria alguma coisa em mente ?</p>

  <p><strong><span style="color:#ff6600;">Luana:</span></strong> Claro! Você pode considerar fazer um piquenique na praia, visitar um museu
  ou fazer uma caminhada na natureza. O que você e sua esposa gostam de fazer juntos?</p>

  <p><strong><span style="color:#3366cc;">Human:</span></strong> Eu e minha esposa gostamos muito de caminhadas.
  Temos uma mochila que carregamos nosso gatinha para todo lado.</p>

  <p><strong><span style="color:#ff6600;">Luana:</span></strong> Parece ótimo! Você pode considerar fazer uma caminhada em um parque local ou reserva natural. Você
  também pode pesquisar trilhas mais longas para fazer em um fim de semana.
  Além disso, você pode trazer um piquenique com você para saborear durante a caminhada.</p>

  <p><strong><span style="color:#3366cc;">Human:</span></strong> Otimas ideias voce tem me dado. O que será que é gostoso levar em um piquenique ?</p>

  <p><strong><span style="color:#ff6600;">Luana:</span></strong> Você pode trazer sanduíches, frutas, vegetais, queijo e biscoitos.
  Você também pode trazer uma garrafa de água ou um refrigerante para beber.
  Além disso, você pode trazer um saco de lixo para descartar resíduos de maneira responsável.</p>

  <p><strong><span style="color:#3366cc;">Human:</span></strong> Olha só que dica bacana. Realmente devo levar uma sacola para nao descartar
  nada na Natureza e assim me divertir com minha esposa.</p>

  <p><strong><span style="color:#ff6600;">Luana:</span></strong> Sim, é importante descartar resíduos de maneira responsável e proteger o meio ambiente.
  Você pode considerar fazer um piquenique em um parque local ou reserva natural onde
  possa descartar resíduos de maneira responsável.</p>

  <p><strong><span style="color:#3366cc;">Human:</span></strong> Perfeito. Muito Obrigado Luana. Otimas dicas. Até a proxima !</p>

  <p><strong><span style="color:#ff6600;">Luana:</span></strong> De nada! Fique em contato se precisar de mais ajuda. Tenha um ótimo fim de semana!</p>
</div>


# [Open Portuguese LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/eduagarcia/open_pt_llm_leaderboard)
Detailed results can be found [here](https://huggingface.co/datasets/eduagarcia-temp/llm_pt_leaderboard_raw_results/tree/main/rhaymison/Mistral-portuguese-luana-7b-chat)

|          Metric          |  Value  |
|--------------------------|---------|
|Average                   |**65.03**|
|ENEM Challenge (No Images)|    59.13|
|BLUEX (No Images)         |    49.24|
|OAB Exams                 |    36.58|
|Assin2 RTE                |    90.47|
|Assin2 STS                |    76.55|
|FaQuAD NLI                |    66.75|
|HateBR Binary             |    77.46|
|PT Hate Speech Binary     |    69.45|
|tweetSentBR               |    59.63|



### Comments

Any idea, help or report will always be welcome.

email: [email protected]

 <div style="display:flex; flex-direction:row; justify-content:left">
    <a href="https://www.linkedin.com/in/rhaymison-cristian-betini-2b3016175/" target="_blank">
    <img src="https://img.shields.io/badge/LinkedIn-0077B5?style=for-the-badge&logo=linkedin&logoColor=white">
  </a>
  <a href="https://github.com/rhaymisonbetini" target="_blank">
    <img src="https://img.shields.io/badge/GitHub-100000?style=for-the-badge&logo=github&logoColor=white">
  </a>
 </div>