Timothy Kassis

I'm having the same issue recently with dc.feat.Mol2VecFingerprint()

``` Failed to featurize datapoint 0, NC(=O)C1=CN(C2OC(COP(=O)(O)OP(=O)(O)OCC3OC(n4cnc5c(N)ncnc54)C(OP(=O)(O)O)C3O)C(O)C2O)C=CC1. Appending empty array Exception message: The vocab attribute was removed from KeyedVector in Gensim 4.0.0. Use KeyedVector's .key_to_index dict, .index_to_key list, and methods...

Here is my code snippet: ```python import deepchem as dc import pandas as pd compounds = pd.read_csv('Targets.csv') featurizer = dc.feat.Mol2VecFingerprint() compounds_list = [] feature_vector_list = [] for index, compound in...

I solved my specific issue by downgrading to gensim 3.8.

Is this related to very slow acceleration (Toyota Corolla) when the lead car accelerates?

Thank you @KevinMusgrave for such a masterpiece of a library. I have a question relating to the VICRegLoss which I think is what is confusing @stereomatchingkiss as well. I read...

Doesn't seem so yet. At least I couldn't find it and it is not in references in the arxiv nor openreview articles

I'd like to see this too. General stats for every drive would be nice. Mainly percent time or miles engaged.

And on a semi-related note, if the line on the map had a different color between engaged vs not engaged. It'll be nice to see that on the map itself.

I believe x-transformers uses the PyTorch attention layer which given certain conditions it defaults to their memory efficient attention implementation. Maybe an unrelated note, but recently I've started having some...