Talia B. Kimber

## Description Check latest CI to see if it's passing. ## Todos Adapt if test not passing - [ ] TODO 1 ## Status - [ ] Ready to go

New questions for a literature review

1

## Description This PR gives more details about the results and model parameters for a potential literature review. See 3 questions below: 1. Can you please give me the lengths...

new dataset

1

## Description This PR would allow to integrate a new data set to the maxsmi workflow. ## Todos - [ ] TBD ## Questions - [ ] Where is the...

**Describe the bug** CI is currently failing on Windows, see [this workflow run.](https://github.com/volkamerlab/maxsmi/actions/runs/3361907830/jobs/5572923329)

CLI

1

## Description This refactor would allow a non-local version of the GitHub repo, only using the conda package. ### TODOs - [ ] Update relative paths to absolute: Currently using...

## Description Update repo with the conda-package. ## Todos - [x] Update readme for installation instructions. - [x] Update readme for acknowledgments for DS. - [x] Add conda-badge - [x]...

Iteration to output folder name.

2

## Description This creates a different output folder name for training the models. Example: instead of generating `output`, it will produce `output_0` by default. The numbering is defined as an...

Adapt code to avoid character replacement, and therefore avoid `new_smiles` in workflow. Similar in the user prediction, remove `new_smiles` column in the csv file.

This augmentation strategy currently runs in a reasonable amount of time on the _free\_solv_ and the _ESOL_small_ data sets. It may be worth considering a subset of the lipophilicity data...

The molecular prediction is currently made by taking the average over associated random SMILES predictions by the model. An added feature could be to take the median instead of the...