molecule-generation icon indicating copy to clipboard operation
molecule-generation copied to clipboard

Implementation of MoLeR: a generative model of molecular graphs which supports scaffold-constrained generation

Results 15 molecule-generation issues
Sort by recently updated
recently updated
newest added

Due to various dependency issues we pin tensorflow to 2.1.0, which is not compatible with Python 3.8+, meaning that only Python 3.7 is supported. It would be great to unpin...

dependencies

The visualisers were recently improved in #10, but several confusing aspects still remain. This issue is meant for tracking all these quirks while they are being cleaned up. - [...

enhancement

We intended to turn off tensorflow warnings (such as those about TensorRT mentioned in #18), but currently that doesn't seem to be working. The problem is likely that we're setting...

bug

Currently [the CI](https://github.com/microsoft/molecule-generation/actions/workflows/ci.yml) does not pin versions for `black` and `flake8`. We should add a pre-commit config (with appropriate instructions in the `README.md`), which could then be used in the...

dependencies

If `MoLeRInferenceServer` receives an invalid SMILES string (e.g. through `molecule_generation encode`), it just hangs, likely because the corresponding process failed, and the server ends up waiting for it indefinitely. This...

bug

Currently CGVAE is not properly tested and not supported by the inference server used by MoLeR. Even though CGVAE works worse than MoLeR across the board, it may still be...

enhancement

There are no tests covering the CLI entry points or the model wrappers, making it easy to break the wrappers during refactoring. While there is one integration test, it's slightly...

enhancement

Hello, thank you for sharing your work. I have a **question** regarding the decoding process. I'm curious about how to decode more than one molecule from a specific latent space,...

question

Hi, Could I check if the script for recreating the evaluation scores as reported in the paper on the Guacamol benchmark is available in this code base? Thanks.

question

with load_model_from_directory(model_dir) as model: embeddings = model.encode(smiles) "WARNING:tensorflow:From ****\tensorflow\python\util\deprecation.py:576: calling function (from tensorflow.python.eager.polymorphic_function.polymorphic_function) with experimental_relax_shapes is deprecated and will be removed in a future version. Instructions for updating: experimental_relax_shapes is...

dependencies