tcr-bert
tcr-bert copied to clipboard
Large language modeling applied to T-cell receptor (TCR) sequences.
I'm using commit `63aa483ec25a8f4727d1905267499ede75f524de` When I run the example command to train PCA-SVM: `python bin/embed_and_train_classifier.py example_files/glanville_np177_training_patient.tsv temp -t example_files/glanville_np177_testing_patients.tsv -c svm -g 0` I get the following: ``` Traceback (most...
Hi Kevin, thanks for creating this model. I am trying to apply it for my dataset on a remote cluster, so I saved tcrbert_model offline as: tcrbert_model = BertModel.from_pretrained("wukevin/tcr-bert") #in...
Hi,sorry to waste your time. But i can't find models located at "/home/wukevin/projects/tcr/tcr_models/bert_class_pird_antigen_cv/bert_class_pird_excl_*".This path is in your computer and it seems that you didn't provide these models.
Thanks for your work. I want to train a new model on new data, could you provide the script or guidance on that? Thanks.
Hi, when I run '**./tcr-bert/jupyter/transformers_glanville_classifier_and_clustering.ipynb**', the TAPE embedding faced some wrong: ``` INFO:root:TAPE command: CUDA_VISIBLE_DEVICES=2 tape-embed unirep /tmp/tmp423qozu0/input.fasta /tmp/tmp423qozu0/output.npz babbler-1900 --tokenizer unirep --seed 1234 Traceback (most recent call last): File...
Thanks for the good work, is it possible to input AA seq with wildcard residues?
`import model_utils # found under the "tcr" folder tcrbert_trb_cls = model_utils.load_classification_pipeline("wukevin/tcr-bert", device=0) df = model_utils.reformat_classification_pipeline_preds(tcrbert_trb_cls([ "C A S S P V T G G I Y G Y T F",...
conda env create -f environment.yml gets this error: Solving environment: failed ResolvePackageNotFound: - blast=2.5.0 - anndata=0.7.5 - scikit-bio=0.5.6 - cudatoolkit=10.2 - muscle=3.8.1551 - louvain=0.7.0 I've tried many ways to manually...