|
--- |
|
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). |
