Update README.md

README.md

@@ -1,80 +1,66 @@
-Math_Arabic_Llama-3.2-3B-Instruct
-Model Description
-Math_Arabic_Llama-3.2-3B-Instruct is a fine-tuned version of the Llama-3.2-3B-Instruct model, specifically optimized for solving mathematical problems in Arabic. This model leverages the power of the Arabic LLaMA Math Dataset to provide accurate and contextually relevant solutions to a wide range of mathematical queries in the Arabic language.
-Key Features
+---
+license: apache-2.0
+tags:
+- llama
+- text-generation
+- instruct
+- arabic
+- math
+- fine-tuned
+datasets:
+- Jr23xd23/Arabic_LLaMA_Math_Dataset
+base_model: meta-llama/Llama-3.2-3B-Instruct
+inference: true
+---
 
-Specialized in Arabic mathematical problem-solving
-Covers a broad spectrum of mathematical topics
-Ideal for educational applications and Arabic-language tutoring systems
-Built on the robust Llama-3.2-3B-Instruct architecture
+# Math_Arabic_Llama-3.2-3B-Instruct
 
-Model Details
+## Model Description
 
-Architecture: Transformer-based language model
-Task: Text Generation (Instruction Following)
-Language: Arabic
-Base Model: meta-llama/Llama-3.2-3B-Instruct
-Dataset: Arabic LLaMA Math Dataset
-Number of Parameters: 3 billion
-Fine-tuned by: Jr23xd23
+**Math_Arabic_Llama-3.2-3B-Instruct** is a fine-tuned version of the **Llama-3.2-3B-Instruct** model, tailored for solving mathematical problems in Arabic. The model was trained using the **[Arabic LLaMA Math Dataset](https://github.com/jaberjaber23/Arabic-LLaMA-Math-Dataset)**, which includes a wide range of mathematical problems in natural language (Arabic). This model is ideal for educational applications, tutoring, and systems that require automatic math problem-solving in Arabic.
 
-Training Data
-The model was fine-tuned on the Arabic LLaMA Math Dataset, which comprises 12,496 diverse mathematical examples covering:
+## Model Details
 
-Basic Arithmetic
-Algebra
-Geometry
-Probability
-Combinatorics
+- **Model Type**: Transformer-based language model fine-tuned for text generation.
+- **Languages**: Arabic
+- **Base Model**: [meta-llama/Llama-3.2-3B-Instruct](https://huggingface.co/meta-llama/Llama-3.2-3B-Instruct)
+- **Dataset**: [Arabic LLaMA Math Dataset](https://github.com/jaberjaber23/Arabic-LLaMA-Math-Dataset)
+- **Number of Parameters**: 3 billion
+- **Fine-tuned by**: [Jr23xd23](https://huggingface.co/Jr23xd23)
 
-Each example in the dataset consists of:
+## Training Data
 
-An Instruction: The problem statement in Arabic
-A Solution: The corresponding answer in Arabic
+The model was fine-tuned on the **Arabic LLaMA Math Dataset**, which consists of 12,496 examples covering various mathematical topics, such as:
+- Basic Arithmetic
+- Algebra
+- Geometry
+- Probability
+- Combinatorics
 
-Intended Use
-Primary Use Cases
+Each example in the dataset includes:
+- **Instruction**: The problem statement in Arabic.
+- **Solution**: The solution to the problem in Arabic.
 
-Solving mathematical problems in Arabic
-Powering educational applications for Arabic-speaking students
-Enhancing Arabic-language tutoring systems
-Facilitating mathematical reasoning tasks in Arabic
+## Intended Use
 
-Usage Example
-Here's how you can use the model with the Hugging Face Transformers library:
-pythonCopyfrom transformers import AutoModelForCausalLM, AutoTokenizer
+### Primary Use Cases:
+- Solving mathematical problems in Arabic
+- Educational applications
+- Tutoring systems for Arabic-speaking students
+- Mathematical reasoning tasks in Arabic
+
+### How to Use
+
+You can use the model in Python with the Hugging Face `transformers` library:
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# Load model and tokenizer
 tokenizer = AutoTokenizer.from_pretrained("Jr23xd23/Math_Arabic_Llama-3.2-3B-Instruct")
 model = AutoModelForCausalLM.from_pretrained("Jr23xd23/Math_Arabic_Llama-3.2-3B-Instruct")
 
 # Example: Solving a math problem in Arabic
 input_text = "ما هو مجموع الزوايا في مثلث؟"  # What is the sum of angles in a triangle?
 inputs = tokenizer(input_text, return_tensors="pt")
-output = model.generate(**inputs, max_length=100)
+output = model.generate(**inputs)
 print(tokenizer.decode(output[0], skip_special_tokens=True))
-Limitations
-
-The model is optimized for mathematical tasks in Arabic and may not perform well on general language tasks.
-Performance may decrease for extremely complex mathematical problems that fall outside the scope of the training dataset.
-The model's responses should be verified for critical applications.
-
-Ethical Considerations
-
-Users should be aware of potential biases in the training data.
-The model should not be used as the sole decision-maker in high-stakes scenarios.
-Implement appropriate safeguards when deploying this model in educational settings.
-
-License
-This model is licensed under the Apache 2.0 License.
-Citation
-If you use this model in your research or projects, please use the following citation:
-bibtexCopy@model{Math_Arabic_Llama_3.2_3B_Instruct,
-  title={Math_Arabic_Llama-3.2-3B-Instruct},
-  author={Jr23xd23},
-  year={2024},
-  publisher={Hugging Face},
-  url={https://huggingface.co/Jr23xd23/Math_Arabic_Llama-3.2-3B-Instruct},
-}
-Acknowledgements
-We extend our gratitude to the creators of the Arabic LLaMA Math Dataset for providing an invaluable resource that made this fine-tuning possible.