mhsotoudeh/astroclip-mini

该数据集包含四个主要特征：image（图像）、spectrum（光谱）、redshift（红移）和targetid（目标ID）。图像特征为152x152x3的浮点数组，光谱特征为7781x1的浮点数组，红移和目标ID分别为浮点数和整数。数据集分为训练集（train）、测试集（test）和验证集（val）三个部分，分别包含1120、240和240个样本。

The dataset includes four main features: image, spectrum, redshift, and targetid. The image feature is a 3D array with a shape of 152x152x3 and a data type of float32. The spectrum feature is a 2D array with a shape of 7781x1 and a data type of float32. The redshift feature is a float32 type value. The targetid feature is an int64 type value. The dataset is divided into three parts: train, test, and val, containing 1120, 240, and 240 samples respectively. The total download size of the dataset is 500187293 bytes, and the total dataset size is 692064000 bytes. The dataset configuration is default, with data file paths being data/train-*, data/test-*, and data/val-* respectively.