Comments (3)
Hello @jorisbalc,
I'd be quite interested in the solution you choose. Can you describe it a bit?
Thanks,
Alan
from bonito.
Hey @touala,
For the training, I used ctc data with called with similar accuracy as regular basecalls. The default in bonito save-ctc-data is .99 if I recall correctly. I assume the reason behind the model trained in deletions is the high accuracy calls in the ctc data contain quite more deletions than mismatches. The lower accuracy ctc data (I used min-save-ctc 0.80) resulted me in a fairly decent model which managed to basecall the modified bases correctly.
from bonito.
Hi @jorisbalc,
Thank you for taking the time to respond. Any other tricks downstream? I've experiment with --min-accuracy-save-ctc
and bonito train --epochs 1 --lr 5e-4 --pretrained [email protected]
but I'm not convinced that this is optimal.
Alan
from bonito.
Related Issues (20)
- Short basecalled R9.4.1 reads using both trained and default models in bonito, but not in guppy
- Bonito Train missing dataset.py HOT 5
- RNA002 model
- RNA004 Does not output any CTC data HOT 3
- bonito basecaller max_reads HOT 1
- Bonito export for dorado error HOT 6
- Feature request: export the list of reads used to train a model
- bonito WASM
- Training data for R10.4, 4 and 5kHz? HOT 3
- Chunks.npy and Dataset.py not being generated HOT 2
- how to best fine tune a model with different references HOT 12
- Applying INT8 quantization, just like in Dorado HOT 2
- How can I set the `gpu_memory_utilization` ?
- bonito duplex runs out of memory HOT 1
- [Errno 2] No such file or directory: '/data/training/ctc-data/dataset.py' HOT 1
- how to choose the model? HOT 1
- Exporting a model - which weights does it use? HOT 2
- Is it possible to use a Bonito trained or fine-tuned model with dorado duplex basecalling.
- Training Data Used for Pre-made models
- Retrieving every paths examined in beam search module
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 bonito.