# Python Training Dataset
This repository contains a pre-tokenized dataset in Apache Arrow format designed for training or fine-tuning language models on Python programming concepts and code.
---
## Dataset Details
* **Dataset Name:** `python-training-dataset`
* **Format:** Apache Arrow (`.arrow`)
* **License:** Python Software Foundation License 2.0 (`psf2`)
* **Data Splits:** Single-shard train/default set (`data-00000-of-00001.arrow`)
---
## Dataset Structure
### Feature Schema
The dataset contains tokenized numerical sequences ready for model ingestion:
| Feature Name | Type | Description |
| --- | --- | --- |
| `input_ids` | `List(int32)` | Tokenized integer IDs representing input sequences
|
### Data Fields
* **`input_ids`**: A list of 32-bit integers representing sequence token IDs generated from a language model tokenizer.
---
## Usage
You can load and inspect this dataset directly using Hugging Face's `datasets` library:
```python
from datasets import …
