jo-mengr/Immune_ALL_human_single

该数据集是一个包含RNA测序数据的测试集，旨在与训练集分离，用于测试。数据来源于cellwhisperer项目和Luecken等的研究，经过处理后转换为Hugging Face数据集格式。数据集可用于训练多模态模型，通过sentence-transformers框架将转录组数据和文本数据进行对齐。数据集中的anndata对象存储在nextcloud上，并提供分享链接以便下载。这些对象包含了使用hvg和pca方法生成的初始嵌入，作为下游模型推理的输入。

This dataset is a test set containing RNA sequencing data, intended to be separate from the training set for testing purposes. The data来源 from the cellwhisperer project and the research by Luecken et al., processed and converted into the Hugging Face dataset format. The dataset can be used to train a multimodal model, aligning transcriptome and text modalities using the sentence-transformers framework. The anndata objects in the dataset are stored on nextcloud, with share links provided for download. These objects contain initial embeddings generated using the hvg and pca methods, serving as inputs for downstream model inference.