aladar/craigslist_bargains

---
license: mit
task_categories:
- text-classification
language:
- en
size_categories:
- 1K<n<10K
---

Processed https://huggingface.co/datasets/craigslist_bargains into a canonical text classification dataset.

The text is the dialogue between the buyer and seller. The label is the category of the product which they are discussing.

Note that some records don't have dialogues, i.e., the `text` field is null.

For an example of a solution to this task, see the notebook [here](https://github.com/kddubey/cappr/blob/main/demos/huggingface/craigslist_bargains.ipynb).

The notebook which processed the original dataset into this one is available [here](https://github.com/kddubey/stackexchange/blob/main/train_on_test_features/bert/craigslist_bargains.ipynb).

原始信息汇总

数据集概述

许可证

• MIT许可证

任务类别

• 文本分类

语言

• 英语

数据规模

• 1K<n<10K

数据描述

• 数据集包含买家和卖家之间的对话文本。
• 标签是他们在讨论的产品类别。
• 注意：部分记录没有对话，即text字段为空。