ivanjaenm/otdata_100K_rowwise_costtype_sqeuclidean_bins30
收藏Hugging Face2025-12-15 更新2025-12-20 收录
下载链接:
https://hf-mirror.com/datasets/ivanjaenm/otdata_100K_rowwise_costtype_sqeuclidean_bins30
下载链接
链接失效反馈官方服务:
资源简介:
---
dataset_info:
features:
- name: source_dist
sequence: float64
- name: target_dist
sequence: float64
- name: input_cost
sequence: float64
- name: input_cost_matrix
sequence:
sequence: float64
- name: input_idx
sequence: float64
- name: input_total
sequence: float64
- name: target_otplan_solve_dense
sequence: float64
- name: target_otplan_solve_sparse
sequence: float64
splits:
- name: train
num_bytes: 20792320000
num_examples: 2400000
- name: test
num_bytes: 5198080000
num_examples: 600000
download_size: 419050670
dataset_size: 25990400000
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
---
数据集信息:
特征项:
- 源分布(source_dist):类型为序列型64位浮点数
- 目标分布(target_dist):类型为序列型64位浮点数
- 输入代价(input_cost):类型为序列型64位浮点数
- 输入代价矩阵(input_cost_matrix):类型为二维序列型64位浮点数
- 输入索引(input_idx):类型为序列型64位浮点数
- 输入总量(input_total):类型为序列型64位浮点数
- 目标最优传输规划稠密解(target_otplan_solve_dense):类型为序列型64位浮点数
- 目标最优传输规划稀疏解(target_otplan_solve_sparse):类型为序列型64位浮点数
数据集划分:
- 训练集(train):字节占用量20792320000,样本总数2400000
- 测试集(test):字节占用量5198080000,样本总数600000
下载总大小:419050670
数据集总占用大小:25990400000
配置项:
- 配置名称:默认(default),数据文件配置:
- 训练集划分:对应数据路径为data/train-*
- 测试集划分:对应数据路径为data/test-*
提供机构:
ivanjaenm



