Moenupa/MSVQA
收藏Hugging Face2026-03-27 更新2026-03-29 收录
下载链接：
https://hf-mirror.com/datasets/Moenupa/MSVQA
下载链接
链接失效反馈官方服务：
资源简介：
---
dataset_info:
- config_name: High_altitude
features:
- name: images
list: image
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: problem
dtype: string
- name: answer
dtype: string
- name: _qid
dtype: int32
- name: _qtype
dtype:
class_label:
names:
'0': counting
'1': classification
'2': true_false
'3': visual_grounding
'4': visual_grounding_fine_grained
splits:
- name: train
num_bytes: 1615623213
num_examples: 5313
- name: test
num_bytes: 471040184
num_examples: 1500
download_size: 2085838862
dataset_size: 2086663397
- config_name: Indoor
features:
- name: images
list: image
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: problem
dtype: string
- name: answer
dtype: string
- name: _qid
dtype: int32
- name: _qtype
dtype:
class_label:
names:
'0': counting
'1': classification
'2': true_false
'3': visual_grounding
'4': visual_grounding_fine_grained
splits:
- name: train
num_bytes: 890153580
num_examples: 5566
- name: test
num_bytes: 160618122
num_examples: 1000
download_size: 1050222197
dataset_size: 1050771702
- config_name: Low_altitude
features:
- name: images
list: image
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: problem
dtype: string
- name: answer
dtype: string
- name: _qid
dtype: int32
- name: _qtype
dtype:
class_label:
names:
'0': counting
'1': classification
'2': true_false
'3': visual_grounding
'4': visual_grounding_fine_grained
splits:
- name: train
num_bytes: 4373576610
num_examples: 10000
- name: test
num_bytes: 781288605
num_examples: 1781
download_size: 5152202002
dataset_size: 5154865215
- config_name: Underwater
features:
- name: images
list: image
- name: conversations
list:
- name: from
dtype: string
- name: value
dtype: string
- name: problem
dtype: string
- name: answer
dtype: string
- name: _qid
dtype: int32
- name: _qtype
dtype:
class_label:
names:
'0': counting
'1': classification
'2': true_false
'3': visual_grounding
'4': visual_grounding_fine_grained
splits:
- name: train
num_bytes: 2788930761
num_examples: 12000
- name: test
num_bytes: 472748322
num_examples: 2000
download_size: 3260098483
dataset_size: 3261679083
configs:
- config_name: High_altitude
data_files:
- split: train
path: High_altitude/train-*
- split: test
path: High_altitude/test-*
- config_name: Indoor
data_files:
- split: train
path: Indoor/train-*
- split: test
path: Indoor/test-*
- config_name: Low_altitude
data_files:
- split: train
path: Low_altitude/train-*
- split: test
path: Low_altitude/test-*
- config_name: Underwater
data_files:
- split: train
path: Underwater/train-*
- split: test
path: Underwater/test-*
---
提供机构：
Moenupa
