seyonechithrananda/bert-loves-chemistry issues and pull requests

#60 - Tokenize incorrect when getting pretrianed feature

Issue - State: open - Opened by Chris-Tang6 about 1 year ago - 5 comments

#59 - Question about chirality and model results from paper

Issue - State: open - Opened by kosonocky over 1 year ago

#59 - Question about chirality and model results from paper

Issue - State: open - Opened by kosonocky over 1 year ago

#58 - Getting SMILES Vector in Pretrained ChemBERTa Model

Issue - State: closed - Opened by ilkersigirci over 1 year ago - 1 comment

#58 - Getting SMILES Vector in Pretrained ChemBERTa Model

Issue - State: closed - Opened by ilkersigirci over 1 year ago - 1 comment

#57 - RuntimeError: Found dtype Double but expected Float

Issue - State: open - Opened by mapengsen about 2 years ago - 3 comments

#57 - RuntimeError: Found dtype Double but expected Float

Issue - State: open - Opened by mapengsen about 2 years ago - 3 comments

#56 - docs: demo, experiments and live inference API on Tiyaro

Pull Request - State: open - Opened by ijonglin over 2 years ago

#55 - Details about curating pubchem dataset

Issue - State: open - Opened by taew0361 over 2 years ago

#54 - Using rawtext

Issue - State: closed - Opened by RoryGarlandGlam over 2 years ago - 1 comment

#53 - notebooks for evaluating finetuning and pretrain scaling

Pull Request - State: closed - Opened by ElanaPearl almost 3 years ago

#52 - Updating Tutorial Link

Pull Request - State: closed - Opened by rbharath almost 3 years ago

#51 - add delete flag to s3 upload

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#50 - Finetune non final ckpts

Pull Request - State: open - Opened by ElanaPearl over 3 years ago

#49 - space issues when finetuning

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#48 - migrate ft utility functions

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#47 - more code for pretraining and finetuning

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#46 - Customization and cleaning

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#45 - finetune with freezing

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#44 - updated environment, fix float issue

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#43 - added float typing

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago - 1 comment

#42 - added support for multiclass classification

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#41 - fixed pruning so doesnt break when training from scratch

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#40 - Add ability to train on MolNet from scratch (without pretraining)

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#39 - made overwrite output dir actually get used

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#38 - only prune state dict if exists

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#37 - Csv finetune

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#36 - change tanh to relu for regression

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#35 - fix regression ft

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#34 - Finetune with pretrained models from HuggingFace Model Hub

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#33 - add methods for generating embeddings

Pull Request - State: closed - Opened by schithranandanurix over 3 years ago

#32 - fixed classification issue; added gitignore

Pull Request - State: closed - Opened by walid0925 over 3 years ago

#31 - Regression text dataset

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#30 - Finetune on multiple datasets

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#29 - hp search

Pull Request - State: closed - Opened by walid0925 over 3 years ago - 2 comments

#28 - New HuggingFace MolNet finetuning script

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#27 - LazyRegressionDataset, compute_norms.py, and fix to loaders

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#26 - Umap

Pull Request - State: closed - Opened by seyonechithrananda over 3 years ago

#25 - Umap notebook for mtg

Pull Request - State: closed - Opened by seyonechithrananda over 3 years ago

#24 - dataset can now handle a list of files

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#23 - added in ability to clean nones and nans on the fly

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#22 - Increase default vocab_size

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#21 - Path issues

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#20 - normalization values are now read from the cache not from the dataset

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#19 - Lazy loading

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#18 - Organize training code

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#17 - Description of Models on Huggingface

Issue - State: open - Opened by TKassis over 3 years ago - 1 comment

#16 - Updates to MLM script

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#15 - Gpu and normalization fixes

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#14 - Elana/multitask gpu

Pull Request - State: closed - Opened by ElanaPearl over 3 years ago

#13 - Add load_best_model_at_end=True

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#12 - Improvements to MLM script

Pull Request - State: closed - Opened by gabegrand over 3 years ago

#11 - Multitask regression

Pull Request - State: closed - Opened by ElanaPearl almost 4 years ago - 2 comments

#10 - Bpe

Pull Request - State: closed - Opened by ElanaPearl almost 4 years ago - 3 comments

#9 - Turn repo into package

Pull Request - State: closed - Opened by ElanaPearl almost 4 years ago

#8 - Pretrain

Pull Request - State: closed - Opened by ElanaPearl almost 4 years ago - 2 comments

#7 - What is the major difference between ChemBERTa_zinc250k_v2_40k and v1 (or others)

Issue - State: open - Opened by yspaik about 4 years ago - 2 comments

#6 - vocab special token - bos_token_id: 0 , overlapping padding index

Issue - State: closed - Opened by yspaik over 4 years ago - 2 comments

#5 - Accuracy metric in the pre-training stage

Issue - State: closed - Opened by yspaik over 4 years ago - 4 comments

#4 - how to use SELFIES with ChemBERTa

Issue - State: closed - Opened by yspaik over 4 years ago - 4 comments

#3 - Create LICENSE

Pull Request - State: closed - Opened by seyonechithrananda over 4 years ago

#2 - Masking multiple tokens at a time

Issue - State: closed - Opened by jannisborn over 4 years ago - 3 comments

#1 - add deepchem msg at end of notebook

Pull Request - State: closed - Opened by seyonechithrananda almost 5 years ago

GitHub / seyonechithrananda/bert-loves-chemistry issues and pull requests