google-deepmind / google-deepmind/deepmind-research

"contribution_scores" (gradient * input) outside the middle 896 bins in Enformer

Open
#306 0 comments 0 reactions 0 assignees View on GitHub
Dominant language
Jupyter Notebook
Stars
15.2k
Forks
2.9k
PR merge metrics
No merged PRs in 30d

Description

Hi,

I appreciate for developing this great tool, Enformer.
I am using it according to the enformer-usage.ipynb.
The length of input sequence is 393,216 bp (128 bp * 3072 bin).
The length of "contribution_scores" is also 393,216.
But, the "contribution_scores" outside the middle part (128 bp * 896 bin) are all zero.
Does this mean the outside part (1088 bin on each side) doesn't contribute to prediction results?
Is there any way to calculate the contribution scores of the outside part?
I am sorry to bother you.

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.