Upload fmri_encoder_3.0model.pth

This is a Transformer-based model trained on multi-modal neuroimaging data, integrating regions of interest (fMRI ROI), age, and gender features for binary classification tasks like autism and ADHD diagnosis. The model leverages datasets such as ABIDE, ADHD-200, and Nilearn’s movie-watching dataset.

The model employs state-of-the-art transformer architectures, pooling mechanisms (mean, max, attention-based), and optimized hyperparameters for robust performance on neuroimaging data.

The model can be used for analyzing functional brain activity and improving diagnostic workflows in neuroimaging research.

fmri_encoder_3.0model.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe32abd187d1bb250aba0935fc3356513ffa91f8584b633839f32ed8e4076c49
+size 450171498