limingcv/Captioned_COCOStuff

---
dataset_info:
features:
- name: image
dtype: image
- name: control_seg
dtype: image
- name: prompt
dtype: string
- name: semantic_seg_map
sequence:
sequence: uint8
- name: panoptic_seg_map
sequence:
sequence: uint8
- name: image_path
dtype: string
splits:
- name: train
num_bytes: 123711791795.25
num_examples: 118287
- name: validation
num_bytes: 5196315752.0
num_examples: 5000
download_size: 61934208857
dataset_size: 128908107547.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
---

The dataset includes multiple features such as images, control segmentation images, prompt texts, semantic segmentation maps, panoptic segmentation maps, and image paths. The dataset is divided into a training set and a validation set, with 118287 and 5000 samples respectively. The total download size of the dataset is 61934208857 bytes, and the total size is 128908107547.25 bytes. The dataset configuration is set to default, with the training and validation data files stored in the data/train-* and data/validation-* paths respectively.