Geneformer / geneformer

13 contributors

History: 111 commits

ctheodoris

update tokenizer to defaults for 95M models for special token and input size

da8cf3d verified about 1 month ago

gene_dictionaries_30m
update with 12L and 20L i4096 gc95M models, multitask and quantiz code about 2 months ago
mtl
pointing dictionaries from the mtl module's init (#397) about 1 month ago
__init__.py

1.22 kB

precommit formatting about 2 months ago
classifier.py

64.7 kB

sync token_dictionary variable name w/ classifier about 1 month ago
classifier_utils.py

23.3 kB

precommit formatting about 2 months ago
collator_for_classification.py

31.2 kB

Refactor: Convert mask_token_id, pad_token_id, and all_special_ids to properties (#395) about 1 month ago
emb_extractor.py

31.8 kB

precommit formatting about 2 months ago
ensembl_mapping_dict_gc95M.pkl

3.96 MB
LFS

Add function for summing of Ensembl IDs (#377) about 2 months ago
evaluation_utils.py

9.75 kB

precommit formatting about 2 months ago
gene_median_dictionary_gc95M.pkl
Detected Pickle imports (2)
- "numpy.core.multiarray.scalar",
- "numpy.dtype"
How to fix it?
1.51 MB
LFS

Add function for summing of Ensembl IDs (#377) about 2 months ago
gene_name_id_dict_gc95M.pkl

2.04 MB
LFS

rename for consistency about 2 months ago
in_silico_perturber.py

65.5 kB

allow model_type valid options to take params model_type : {"Pretrained", "GeneClassifier", "CellClassifier", "MTLCellClassifier", "MTLCellClassifier-Quantized"} (#390) about 1 month ago
in_silico_perturber_stats.py

44.9 kB

precommit formatting about 2 months ago
mtl_classifier.py

13.7 kB

"save_model_without_heads" is redundant (#385) about 1 month ago
perturber_utils.py

33 kB

precommit formatting about 2 months ago
pretrainer.py

36.6 kB

update with 12L and 20L i4096 gc95M models, multitask and quantiz code about 2 months ago
token_dictionary_gc95M.pkl

426 kB
LFS

update with 12L and 20L i4096 gc95M models, multitask and quantiz code about 2 months ago
tokenizer.py

27.3 kB

update tokenizer to defaults for 95M models for special token and input size about 1 month ago

Detected Pickle imports (2)