Embryoid Bodies 2018

Embryoid Bodies 2018 dataset.

Constants

DataModule

EmbryoidBodies2018DataModule

 EmbryoidBodies2018DataModule (time_col:str='samples',
                               data_dir:str='/home/runner/Downloads/embryo
                               id_2018', batch_size:int=32,
                               use_time_dataset:bool=True,
                               primary:str='counts',
                               dl_kwargs:dict[str,typing.Any]=<factory>,
                               perc_train:float=0.7, perc_valid:float=0.1,
                               perc_test:float=0.2)

eb = EmbryoidBodies2018DataModule(
    primary='pca',
    batch_size=8,
)

eb.does_data_dir_exists, eb.does_sc_rna_dir_exists, eb.has_all_timepoint_dirs

(True, True, True)

eb.setup()

Loading data
Data ready!

dl = eb.train_dataloader()

eb.train_ds.df[eb.train_ds.df.columns[-5:]].head()

	d97	d98	d99	d100	samples
5100	0.881895	-1.463558	0.309755	-0.156539	Day 06-09
15125	1.099650	-1.794608	-1.442124	-0.621449	Day 24-27
9431	-1.441780	0.068481	2.492943	2.289345	Day 12-15
13078	-0.047084	0.543089	1.502882	1.567946	Day 18-21
3193	0.652287	0.758303	0.681693	-0.214407	Day 06-09

eb.train_ds.t_min, eb.train_ds.t_max

(0, 4)

for x, y in dl:
    break
x.shape, y.shape

(torch.Size([8, 5, 100]), torch.Size([8, 5]))

for x, y in eb.train_ds:
    break
x.shape, y.shape

(torch.Size([5, 100]), torch.Size([5]))