swansonk14 / admet_ai Goto Github PK
View Code? Open in Web Editor NEWTraining and prediction scripts for Chemprop models trained on ADMET datasets
License: MIT License
Training and prediction scripts for Chemprop models trained on ADMET datasets
License: MIT License
Hi Kyle,
Sorry to bother you again, but on my local installation the maximum number of molecules to predict in one run is set to 1000. How can I increase this number? Where is this limit defined?
Also, it would be very useful if one could include an ID column in a csv file to identify molecules in the results. Now you have rely on the smiles only as it seems (?)
Thanks/Evert
Hello, for some compounds your tool predicts negative hepatocyte clearance values. From a practical standpoint that doesnt make much sense. So, should I just interpret these as 0 values (i.e. no clearance)?
Hello,when I run the inference.py, I get this problem. Can you help me solve the problem?
FileNotFoundError: [Errno 2] No such file or directory: '/home/xuhu/project/admet_ai/admet_ai/resources/data/reference_no_kd_ic50.csv'
Hello, thank you for the nice programmatic interface to the models.
Would it be possible to add a couple functions to the Model class to allow it to make predictions by passing it the Mol object directly or FP and descriptors?
I ask because in larger projects it is quite common to have the Mol object stores somewhere anyway and in this way I would not have to keep creating such object over and over.
Since I'll have to write a wrapper around the library I'll probably do it anyway, but it would be nice to have the functionality directly in the library.
...
admet_predict in Cli Mode throw error:Reindexing only valid with uniquely valued Index objects, what's the reason and how fix it
Hi Kyle,
Thanks for making this great tool available to the public. By default the web server uses port 5000, but is there a way to change this, as many web tools by default use this port. Where is this configured?
Much appreciated/Evert
Hey, thanks for putting out such a great package!
I'm struggling with installing this into an environment with a number of other packages - is it 100% necessary to pin every requirement to the exact version in requirements.txt
?
chemfunc==1.0.4
chemprop==1.6.1
flask==3.0.0
gunicorn==21.2.0
numpy==1.25.2
openpyxl==3.1.2
pandas==2.1.0
PyTDC==0.4.1
rdkit==2023.3.3
seaborn==0.13.0
tqdm==4.66.1
typed-argument-parser==1.9.0
In particular, the numpy
/pandas
/rdkit
specifications seem pretty tight, particularly in light of standard semantic versioning.
Hi Kyle, thanks so much for putting this togther. I have encounterd a "Depreciation warning: please use MorganGenerator." as a repeated error message when running the local python version.
I think this RDKit blog covers the problem https://greglandrum.github.io/rdkit-blog/posts/2023-01-18-fingerprint-generator-tutorial.html
All the best
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.