GigaSSL-TCGA_embeddings-Models

Published: 21 November 2023| Version 3 | DOI: 10.17632/d573xfd9fg.3
Contributor:
Tristan Lazard

Description

You will find here the models needed to compute the Giga-SSL embeddings of WSIs. You will also find the Giga-SSL embeddings of the full TCGA-FFPE WSI dataset, computed with two models: one using a MoCo encoder, the second using the CTransPath tile-encoder. To load the embeddings: """python import numpy as np dico_embeddings = np.load('./Embeddings/TCGA-gigassl_moco.npy', allow_pickle=True).item() """

Files

Institutions

Mines ParisTech

Categories

Cancer, Cancer Cell

Licence