Thanks for your great work! Can you share how to implemented t-SNE visualization of the high-level embeddings from the last layer of the image encoder on CheXpert 5x200？

QtacierP / PRIOR

Official repository for the paper "Prototype Representation Joint Learning from Medical Images and Reports, ICCV 2023".

Apache License 2.0

62 stars 5 forks source link

Thanks for your great work! Can you share how to implemented t-SNE visualization of the high-level embeddings from the last layer of the image encoder on CheXpert 5x200？ #9

Open Dramaeyp opened 6 months ago

QtacierP commented 6 months ago

Thanks for your attention. We utilize sklearn.manifold.TSNE to implement the T-SNE process. We use the last layer's output from the image encoder as the features. Maybe this question is a good example.