precommit formatting

update with 12L and 20L i4096 gc95M models, multitask and quantiz code

add typing list import

move dicts to init

update get_embs with token_gene_dict arg

fix cell state gene embeddings bug (#345)

add load model for train and fix validate anchor gene error

Handle case of single gene del for isp modeling of gene embs

edit docstring codeblock highlighting

change doc formatting

add sphinx docs

update dependencies, reinstate compatibility with python<3.9 with typing for List

Add functions for extracting gene embeddings, move state_embs_dict outside isp, fix bugs in isp

Add option to output embs as tensor

Fixed bugs related to overexpressing genes (#229)

Fixed bug with the double removing of indices when cell_states_to_model is false (#188)

Upload in_silico_perturber.py (#187)

Added feature to perturb a set of indices to help with debugging and with very large runtimes (#175)

Remove print statement from PR

Fixed bug in gen_attention_mask with len > max_len (#158)

Fixed error with perturbing individual genes and updated ways to specify cell_states_to_model (#146)

Add function to extract and plot cell embeddings

Fix min_genes to be >= tokens to perturb as a group

Update isp to allow modeling single perturbation in multiple cells as batches

Update internal format of anchor token to list for consistency with genes to perturb

Add filtering for start state cells prior to in silico perturbation when modeling cell states

Correct order of state dict in in silico perturber stats and tensor dims of alt state emb in in silico perturber

Modify quant_layers to convert layer nums to integers before calculating max

Modify documentation for modeling only 2 cell states

Add instructions for modeling only 2 states and modify stats script for that option

Add function to create remainder emb for in silico overexpression batch

Fix bug in clearing cache

Fix filter_data to allow value of None for no filtering

add in silico perturbation module