File size: 4,073 Bytes
5032722 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 | #!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Created on Wed Jan 29 17:36:42 2025
About:
=======
initialize dataloaders and pytorch datasets
"""
import pickle
import torch
from torch.utils.data import DataLoader
import numpy as np
from dloaders.CountsDset import CountsDset
from dloaders.CountsDset import jax_collator as collator
from dloaders.init_dataloader import init_dataloader
from dloaders.init_time_array import init_time_array
def init_counts_dset( args,
task,
training_argparse=None,
include_dataloader=True ):
"""
initialize the dataloaders
"""
#################################
### training-specific options #
#################################
if task in ['train',
'resume_train']:
only_test = False
t_per_sample = args.pred_config['times_from'] == 't_per_sample'
t_array_for_all_samples = init_time_array(args)
subs_only = (args.pred_config['indel_model_type'] is None)
if args.pred_config['subst_model_type'].lower() == 'hky85':
emission_alphabet_size = 4
else:
emission_alphabet_size = 20
#############################
### eval-specific options #
#############################
elif task in ['eval']:
only_test = True
t_per_sample = training_argparse.pred_config['times_from'] == 't_per_sample'
t_array_for_all_samples = init_time_array(training_argparse)
subs_only = (training_argparse.pred_config['indel_model_type'] is None)
if training_argparse.pred_config['subst_model_type'].lower() == 'hky85':
emission_alphabet_size = 4
else:
emission_alphabet_size = 20
#################
### LOAD DATA #
#################
# test data
assert type(args.test_dset_splits) == list
print('Test dset:')
for s in args.test_dset_splits:
print(s)
print()
test_dset = CountsDset( data_dir = args.data_dir,
split_prefixes = args.test_dset_splits,
emission_alphabet_size = emission_alphabet_size,
t_per_sample = t_per_sample,
subs_only = subs_only,
toss_alignments_longer_than = args.toss_alignments_longer_than)
out = {'test_dset': test_dset,
't_array_for_all_samples': t_array_for_all_samples}
# training data
if not only_test:
assert type(args.train_dset_splits) == list
print('Training dset:')
for s in args.train_dset_splits:
print(s)
print()
training_dset = CountsDset( data_dir = args.data_dir,
split_prefixes = args.train_dset_splits,
emission_alphabet_size = emission_alphabet_size,
t_per_sample = t_per_sample,
subs_only = subs_only,
toss_alignments_longer_than = args.toss_alignments_longer_than)
out['training_dset'] = training_dset
############################################
### create dataloaders, output dictionary #
############################################
if include_dataloader:
test_dl = init_dataloader(args = args,
pytorch_custom_dset = test_dset,
shuffle = False,
collate_fn = collator)
out['test_dl'] = test_dl
if not only_test:
training_dl = init_dataloader(args = args,
pytorch_custom_dset = training_dset,
shuffle = True,
collate_fn = collator)
out['training_dl'] = training_dl
return out
|