guopengf / FL-MRCM

Multi-institutional Collaborations for Improving Deep Learning-based Magnetic Resonance Image Reconstruction Using Federated Learning
MIT License
45 stars 10 forks source link

how to plot latent features #4

Closed kingaza closed 3 years ago

kingaza commented 3 years ago

Hello,

The work is very interesting! I am familar with MRI reconstruction, but not with federated learning. Could you provide me some hints on ploting letent feature distribution from different datasets, as shown in Figure 1 d) and e) Thanks a lot.

Best Regards, Kingaza

guopengf commented 3 years ago

Hi Kingaza, I use sklearn.manifold.TSNE. Please find more info at https://scikit-learn.org/stable/modules/generated/sklearn.manifold.TSNE.html.

kingaza commented 3 years ago

I actually know TSNE. But I don't now how to do this dimension reduction from which features to the 2 features...

guopengf commented 3 years ago

I see your problem. What I did was simple. I just reshape the each whole image to 1-dimension vector, then stack them together.