yahoo-answers / README.md
tomaarsen's picture
tomaarsen HF staff
Update README.md
93b3605
|
raw
history blame
No virus
6.36 kB
---
language:
- en
multilinguality:
- monolingual
size_categories:
- 1M<n<10M
task_categories:
- feature-extraction
- sentence-similarity
pretty_name: Yahoo Answers
tags:
- sentence-transformers
dataset_info:
- config_name: question-answer-pair
features:
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 441860501
num_examples: 681164
download_size: 296974225
dataset_size: 441860501
- config_name: title-answer-pair
features:
- name: title
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 532353635
num_examples: 1198260
download_size: 359777740
dataset_size: 532353635
- config_name: title-question-answer-pair
features:
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 462195629
num_examples: 599417
download_size: 308542541
dataset_size: 462195629
- config_name: title-question-pair
features:
- name: title
dtype: string
- name: questions
dtype: string
splits:
- name: train
num_bytes: 190935497
num_examples: 659896
download_size: 132675030
dataset_size: 190935497
configs:
- config_name: question-answer-pair
data_files:
- split: train
path: question-answer-pair/train-*
- config_name: title-answer-pair
data_files:
- split: train
path: title-answer-pair/train-*
- config_name: title-question-answer-pair
data_files:
- split: train
path: title-question-answer-pair/train-*
- config_name: title-question-pair
data_files:
- split: train
path: title-question-pair/train-*
---
# Dataset Card for Yahoo Answers
This dataset is a collection of pairs containing titles, questions, and answers collected from Yahoo Answers. See the [Yahoo Answers](https://www.kaggle.com/datasets/soumikrakshit/yahoo-answers-dataset) dataset for additional information. This dataset can be used directly with Sentence Transformers to train embedding models.
## Dataset Subsets
### `title-question-answer-pair` subset
* Columns: "question", "answer"
* Column types: `str`, `str`
* Examples:
```python
{
'question': "why doesn't an optical mouse work on a glass table? or even on some surfaces?",
'answer': "why doesn't an optical mouse work on a glass table? Optical mice use an LED and a camera to rapidly capture images of the surface beneath the mouse. The infomation from the camera is analyzed by a DSP (Digital Signal Processor) and used to detect imperfections in the underlying surface and determine motion. Some materials, such as glass, mirrors or other very shiny, uniform surfaces interfere with the ability of the DSP to accurately analyze the surface beneath the mouse. \\nSince glass is transparent and very uniform, the mouse is unable to pick up enough imperfections in the underlying surface to determine motion. Mirrored surfaces are also a problem, since they constantly reflect back the same image, causing the DSP not to recognize motion properly. When the system is unable to see surface changes associated with movement, the mouse will not work properly.",
}
```
* Collection strategy: Reading the `title-answer-pair` and `title-question-pair` datasets, matching up the titles, filtering on just 1 question and 1 answer, and then concatenating the title + the question as the question.
* Deduplified: No
### `title-answer-pair` subset
* Columns: "title", "answer"
* Column types: `str`, `str`
* Examples:
```python
{
'title': "why doesn't an optical mouse work on a glass table?",
'answer': 'Optical mice use an LED and a camera to rapidly capture images of the surface beneath the mouse. The infomation from the camera is analyzed by a DSP (Digital Signal Processor) and used to detect imperfections in the underlying surface and determine motion. Some materials, such as glass, mirrors or other very shiny, uniform surfaces interfere with the ability of the DSP to accurately analyze the surface beneath the mouse. \\nSince glass is transparent and very uniform, the mouse is unable to pick up enough imperfections in the underlying surface to determine motion. Mirrored surfaces are also a problem, since they constantly reflect back the same image, causing the DSP not to recognize motion properly. When the system is unable to see surface changes associated with movement, the mouse will not work properly.',
}
```
* Collection strategy: Reading the Yahoo Answers (title, answer) dataset from [embedding-training-data](https://huggingface.co/datasets/sentence-transformers/embedding-training-data).
* Deduplified: No
### `title-question-pair` subset
* Columns: "title", "question"
* Column types: `str`, `str`
* Examples:
```python
{
'title': "why doesn't an optical mouse work on a glass table?",
'questions': 'or even on some surfaces?',
}
```
* Collection strategy: Reading the Yahoo Answers (title, question) dataset from [embedding-training-data](https://huggingface.co/datasets/sentence-transformers/embedding-training-data).
* Deduplified: No
### `question-answer-pair` subset
* Columns: "question", "answer"
* Column types: `str`, `str`
* Examples:
```python
{
'question': 'or even on some surfaces?',
'answer': 'Optical mice use an LED and a camera to rapidly capture images of the surface beneath the mouse. The infomation from the camera is analyzed by a DSP (Digital Signal Processor) and used to detect imperfections in the underlying surface and determine motion. Some materials, such as glass, mirrors or other very shiny, uniform surfaces interfere with the ability of the DSP to accurately analyze the surface beneath the mouse. \\nSince glass is transparent and very uniform, the mouse is unable to pick up enough imperfections in the underlying surface to determine motion. Mirrored surfaces are also a problem, since they constantly reflect back the same image, causing the DSP not to recognize motion properly. When the system is unable to see surface changes associated with movement, the mouse will not work properly.',
}
```
* Collection strategy: Reading the Yahoo Answers (question, answer) dataset from [embedding-training-data](https://huggingface.co/datasets/sentence-transformers/embedding-training-data).
* Deduplified: No