Doub7e/SD-CLIP-alignment-composition-T5

---
dataset_info:
features:
- name: image
dtype: image
- name: prompt
dtype: string
- name: clip_pred
dtype: string
- name: T5_last_hidden_states
sequence:
sequence:
sequence: float32
splits:
- name: train
num_bytes: 431432487.0
num_examples: 900
download_size: 419405907
dataset_size: 431432487.0
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
---
# Dataset Card for "SD-CLIP-alignment-composition-T5"

[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)

原始信息汇总

数据集概述

数据集信息

特征

• image: 图像数据
• prompt: 字符串
• clip_pred: 字符串
• T5_last_hidden_states: 多层嵌套的浮点数序列

分割

• train: 训练集
  • 字节数: 431432487.0
  • 样本数: 900

大小

• 下载大小: 419405907
• 数据集大小: 431432487.0

配置

• default: 默认配置
  • 数据文件:
    • train: data/train-*