michsethowusu/Code-170k-bambara

Dataset Description

Code-170k-bambara is a groundbreaking dataset containing 176,999 programming conversations, originally sourced from glaiveai/glaive-code-assistant-v2 and translated into Bambara, making coding education accessible to Bambara speakers.

🌟 Key Features

• 176,999 high-quality conversations about programming and coding
• Pure Bambara language - democratizing coding education
• Multi-turn dialogues covering various programming concepts
• Diverse topics: algorithms, data structures, debugging, best practices, and more
• Ready for instruction tuning of Large Language Models

🎯 Use Cases

• Training Bambara-language coding assistants
• Building educational tools for Bambara developers
• Researching multilingual code generation
• Creating programming tutorials in Bambara
• Supporting low-resource language AI development

Dataset Structure

Data Fields

• conversations: A list of conversation turns, where each turn contains:
  • from: The speaker ("human" or "gpt")
  • value: The message content in Bambara

Example

{
  "conversations": [
    {
      "from": "human",
      "value": "[Question in Bambara]"
    },
    {
      "from": "gpt",
      "value": "[Answer in Bambara]"
    }
  ]
}

Usage

Loading the Dataset

from datasets import load_dataset

# Load the dataset
dataset = load_dataset("michsethowusu/Code-170k-bambara")

# Access training data
train_data = dataset['train']

# Example: Print first conversation
for turn in train_data[0]['conversations']:
    print(f"{turn['from']}: {turn['value']}")

Citation

@dataset{code170k_bambara,
  title={Code-170k-bambara: Programming Conversations in Bambara},
  year={2025},
  publisher={Hugging Face},
  url={https://huggingface.co/datasets/michsethowusu/Code-170k-bambara}
}

License

This dataset is released under the Apache 2.0 License.

---

Thank you for using Code-170k-bambara to advance programming education in Bambara! 🌍✨