Aaditya Prasad

icon indicating a website link https://aaprasad.github.io

icon company MIT Brain & Cognitive Sciences Dept icon location Cambridge, MA icon location Phd Student @ MIT Brain & Cognitive Sciences Dept. Prev: Data Science MS + Bioinfo BS @ UCSD w @talmolab & @ucsdmanorlab. Interested in Deep Learning x Neuro

Results 5 comments of Aaditya Prasad

BatchNorm error when calling `model = sm.Unet(BACKBONE, encoder_weights='imagenet')`

did you ever figure out a solution to this problem?

Pre-Averaged Filtered Data

Yes! the size isn't an issue for me but i wanted to calculate my own RDMs using crossnobis distance measures for which I believe I need the trial data as...

Pre-Averaged Filtered Data

alternatively the trial averaged responses would actually be preferable if possible ie a tensor of shape Num Trials x Num Images x Num Neurons but either is fine

PermissionError with ModelCheckpoints

Hi, yes it does. I'm currently using: ```albumentations 1.3.1 pyhd8ed1ab_0 conda-forge cuda-cudart 12.1.105 0 nvidia cuda-cupti 12.1.105 0 nvidia cuda-libraries 12.1.0 0 nvidia cuda-nvrtc 12.1.105 0 nvidia cuda-nvtx 12.1.105 0...

A couple of issues in model_predictions from submission.py

Yeah I used my own model!