Upload README.md with huggingface_hub

README.md

@@ -1,21 +1,67 @@
 ---
-dataset_info:
-  features:
-  - name: system
-    dtype: string
-  - name: instruction
-    dtype: string
-  - name: response
-    dtype: string
-  splits:
-  - name: train
-    num_bytes: 23720
-    num_examples: 6
-  download_size: 28720
-  dataset_size: 23720
-configs:
-- config_name: default
-  data_files:
-  - split: train
-    path: data/train-*
+annotations_creators:
+- no-annotation
+language:
+- en
+license: mit
+multilinguality:
+- monolingual
+size_categories:
+- n<1K
+task_categories:
+- sequence-modeling
+task_ids:
+- language-modeling
 ---
+
+# burme-coder-max
+
+A coder-focused dataset designed for training AI coding assistants with strong programming capabilities.
+
+## Dataset Description
+
+This dataset contains coding examples with a comprehensive system prompt that enables AI models to:
+- Write clean, well-documented code
+- Follow best practices and design patterns
+- Provide working solutions with explanations
+- Handle edge cases appropriately
+- Offer alternative approaches
+
+## Dataset Structure
+
+| Field | Type | Description |
+|-------|------|-------------|
+| system | string | System prompt defining coder behavior |
+| instruction | string | User's coding task or question |
+| response | string | Complete code solution with explanations |
+
+## Usage
+
+```python
+from datasets import load_dataset
+
+dataset = load_dataset("amkyawdev/burme-coder-max")
+train_data = dataset['train']
+
+# Access an example
+example = train_data[0]
+print(example['instruction'])
+print(example['response'])
+```
+
+## Categories Covered
+
+- Algorithm & Data Structures
+- Web Development (Node.js, Express)
+- Database & SQL
+- TypeScript & Type Safety
+- Go & Concurrent Programming
+- Python Best Practices
+
+## Version
+
+1.0.0
+
+## License
+
+MIT License