bartmch

How to add (source to target) causality between two 1D time series? (without using numpy tril)

1

Is there a better (computationally cheaper) way to quantify the relationship between a source>target time series instead of setting the lower triangle matrix to inf after computing it? ``` #...

Hi, I understand that we need to scale the learning rate in DDP to make sure the gradients are averaged correctly at the end. But I'm confused about the choice...

RISE returns different heatmap for same input

7

Hey, after making your model torch>1.0 compatible I got the RISE "explainer" predictions for the covid-19 file "4-x-day1.jpg" twice. The predictions are the same:   but the heatmaps produced...

1. Rename torch1.0 layers. 2. Make it possible to run when torch.cuda isn't available (CPU)

I am currently using: ``` import sys import os import pkgutil import pkg_resources egg_path = '/path/to/downloaded/checkpoint.egg' sys.path.insert(0, egg_path) try: import checkpoint print(dir(checkpoint)) print(checkpoint.__path__) # Iterate through all modules in the...

I am looking into truncating the input tokens such that it fits the model max tokens. Because I don't want to do unnecessary tokenisation on the OWUI side all the...