Comments (4)
Hi Jeff,
The training code takes a parameter called "batch_buffer" set to 4096 for these models that tells the loss function to ignore the 4096 bp on each side. The 960 length is taken after slicing these positions out.
The target order won't match that in the supplementary table. Where are you getting this model from? I'm not currently saving the target labels with the model; I usually have the training dataset nearby.
Best,
Dave
from basenji.
That makes sense, thank you.
I got it from kipoi, a genomics model zoo that came out around March. Specifically, here.
If the output doesn't match the supplementary table, could you provide a mapping?
Jeff
from basenji.
Got ya. Then it should refer to https://storage.googleapis.com/131k/sample_wigs.txt as specified in https://github.com/calico/basenji/blob/master/manuscript/get_data.sh.
from basenji.
Cool, i'll direct others using kipoi to do the same.
Much appreciated,
Jeff
from basenji.
Related Issues (20)
- distiller HOT 6
- No checkpoints found. HOT 1
- Problem with calculating the target based on the gene sequence HOT 1
- None in maps HOT 4
- .hic
- Expose 'raw' as an argument? HOT 2
- Black list HOT 6
- handling blacklist\unmappable HOT 1
- basenji_data_align.py train/test/validation splits HOT 2
- How to set different GPU No. ? HOT 2
- Stable environment configuration HOT 1
- How to convert the keras model to a basenji model HOT 1
- Pre-process tutorial file not working HOT 2
- tutorial training with --sample 0.05 HOT 3
- Making Akita predictions for different bin size HOT 2
- Attention layer HOT 4
- Seeking Guidance on bam_cov.py Parameters HOT 2
- When using Akita for single-cell prediction on the HCT116 cell line, the Pearson correlation coefficient is much lower. HOT 8
- Akita: Loading weights for newer models results in mismatch HOT 2
- Issue with final matrix?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from basenji.