def test_get_deepnog_root():
deepnog_root = get_deepnog_root()
# Ensure this is the correct directory will all the subpackages
subpackages = [
'client',
'config',
'data',
'learning',
'models',
'tests',
'utils',
]
for pkg in subpackages:
d = deepnog_root / pkg
assert d.is_dir()
# Also check presence of some files
some_files = [
"tests/data/test_inference_short.csv",
"tests/parameters/test_deepnog.pthsmall",
"config/deepnog_config.yml",
]
for file_ in some_files:
f = deepnog_root / file_
assert f.is_file()
# Ensure certain things are not present in deepnog
unexpected_files = [
"me/no/think.so",
"plagiarism/manuscript.tex",
"bugs",
]
for file_ in unexpected_files:
f = deepnog_root / file_
assert not f.exists()
"""
from itertools import repeat
from functools import partial
import pytest
import numpy as np
from pandas import read_csv
from sklearn.preprocessing import LabelEncoder
from torch.utils.data import DataLoader
from Bio.Seq import Seq
from Bio.SeqRecord import SeqRecord
from deepnog.data import dataset as ds
from deepnog.tests.utils import get_deepnog_root
TESTS = get_deepnog_root() / "tests"
test_file = TESTS / "data/GCF_000007025.1.faa"
test_file_gzip = TESTS / "data/GCF_000007025.1.faa.gz"
TRAINING_FASTA = TESTS / "data/test_training_dummy.faa"
TRAINING_LABELS = TESTS / "data/test_training_dummy.faa.csv"
EXPECTED_IDS_WITH_LABEL = [f'test_all_A{x}' for x in range(11)] \
+ [f'test_all_C{x}' for x in range(11)] \
+ [f'M{x:02d}' for x in range(1, 9)]
EXPECTED_IDS = [f'test_all_A{x}' for x in range(12)] \
+ [f'test_all_C{x}' for x in range(12)] \
+ [f'M{x:02d}' for x in range(1, 11)]
LABELS_WRONG_COL_NAMES = TESTS / "data/test_inference_short_wrong_column_names.csv"
@pytest.mark.parametrize("f", [
test_file,
import pytest
import numpy as np
import pandas as pd
from deepnog.data import train_val_test_split, group_train_val_test_split
from deepnog.tests.utils import get_deepnog_root
DEEPNOG_ROOT = get_deepnog_root()
TESTS = DEEPNOG_ROOT/"tests"
DATAFRAME_GROUP = TESTS/"data/test_split.csv"
DATAFRAME_SIMPLE = TESTS/"data/test_simple_split.csv"
@pytest.mark.parametrize('ratio', [[.4, .3, .3], [4, 3, 3]])
def test_simple_split(ratio):
df = pd.read_csv(DATAFRAME_SIMPLE)
train, val, test = ratio
res = train_val_test_split(df,
train_ratio=train,
validation_ratio=val,
test_ratio=test,
random_state=123,
stratify=True,
shuffle=True,
verbose=0)
for group in [res.uniref_train, res.uniref_val, res.uniref_test]:
assert group is None
assert res.X_train.shape == (4, )
assert res.X_val.shape == (3, )
from pathlib import Path
import tempfile
import pytest
import numpy as np
from deepnog.learning import fit
from deepnog.tests.utils import get_deepnog_root
DEEPNOG_TESTS = get_deepnog_root() / "tests"
TRAINING_FASTA = DEEPNOG_TESTS / "data/test_training_dummy.faa"
TRAINING_CSV = DEEPNOG_TESTS / "data/test_training_dummy.faa.csv"
Y_TRUE = np.array([[
0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 1, 1, 1,
1, 1, 1, 1, 1
],
[
0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0, 2, 0,
2, 0, 2, 1, 1, 1, 1, 1, 1, 1, 1
]])
@pytest.mark.parametrize('batch_size', [
4,
])
@pytest.mark.parametrize('num_workers', [0, 2])
def test_shuffled_training(batch_size, num_workers):
results = fit(architecture='deepnog',
module='deepnog',
cls='DeepNOG',
training_sequences=TRAINING_FASTA,
validation_sequences=TRAINING_FASTA,