Arun

I am closing this pull request as I am not able to prioritize time for this, sorry!

It will also be great if item names in chart of account could be renamed as well as if regrouping is allowed.

Hey, thanks for reporting. Here is a smaller example to reproduce: ```py import deepchem as dc import pandas as pd featurizer = dc.feat.RobertaFeaturizer.from_pretrained('seyonec/SMILES_tokenized_PubChem_shard00_160k') df = pd.DataFrame({'smiles': ['CCC', 'CC(=O)CCC']}) result =...

@rbharath The featurizer is not used by any model. As @quincylin1 suggested, a solution would be to convert it to a list if it is a `pd.DataFrame` in the `_featurize`...

It mostly looks good. Maybe you need a type check for `datapoints`. If `datapoints` is already a list, then `tolist` call will fail on `datapoints`. Also, you can directly pass...

I don't think so. Could you try importing rdkit alone as `from rdkit.Chem import AllChem`? Because the error seems to arise form rdkit.

At this point, I think generation cannot be done. But MLM can be done. In the initialization of hugging-face model, along with model and tokenizer, I would suggest also to...

For GraphConvModel, I think `ConvMolFeaturizer` should be used instead of `CircularFingerprint`. Could you retry the same with ConvMolFeaturizer?

@angelinalg just a gentle reminder

Hey, it seems there are some failing tests (see [here](https://github.com/deepchem/deepchem/actions/runs/8299238434/job/22714480583?pr=3898#step:14:804)). Could you take a look at it?