Goal: load model and namespace my_func into namespace for pytest run.
The model and its namespace are fine in project code runtime.
AttributeError only occurs during pytest run.
Placing my_func() anywhere in test script doesn't prevent error.
tests/test_predict.py:
import pytest
from fastai.vision.all import *
from project.predict import my_func, prediction
model = load_learner('model_path/', cpu=True)
model.load(fast_ai_params.weights)
def my_test():
pred = prediction(tile_params, model, full_tile_path)
assert pred.shape
This test script runs fine statically, via. python3. Indicating that both the model and namespace are loaded in, without errors.
(venv) me@ubuntu-pcs:~/PycharmProjects/project$ python3 tests/test_predict.py
(venv) me@ubuntu-pcs:~/PycharmProjects/project$
However, when I run python -m pytest tests/test_predict.py:
=========================================================================================================================== ERRORS ============================================================================================================================
___________________________________________________________________________________________________________ ERROR collecting tests/test_predict.py ____________________________________________________________________________________________________________
tests/test_predict.py:16: in <module>
model = load_learner(fast_ai_params.model, cpu=fast_ai_params.cpu)
../../miniconda3/envs/venv/lib/python3.9/site-packages/fastai/learner.py:414: in load_learner
try: res = torch.load(fname, map_location=map_loc, pickle_module=pickle_module)
../../miniconda3/envs/venv/lib/python3.9/site-packages/torch/serialization.py:594: in load
return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)
../../miniconda3/envs/venv/lib/python3.9/site-packages/torch/serialization.py:853: in _load
result = unpickler.load()
E AttributeError: Custom classes or functions exported with your `Learner` are not available in the namespace currently.
E Please re-declare or import them before calling `load_learner`:
E Can't get attribute 'my_func' on <module 'pytest.__main__' from '/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/pytest/__main__.py'>
====================================================================================================================== warnings summary =======================================================================================================================
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional_pil.py:387
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional_pil.py:387: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
def resize(img, size, interpolation=Image.BILINEAR):
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional_pil.py:545
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional_pil.py:545: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
def perspective(img, perspective_coeffs, interpolation=Image.BICUBIC, fill=None):
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional.py:288
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional.py:288: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
def resize(img: Tensor, size: List[int], interpolation: int = Image.BILINEAR) -> Tensor:
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional.py:419
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/functional.py:419: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
img: Tensor, top: int, left: int, height: int, width: int, size: List[int], interpolation: int = Image.BILINEAR
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:26
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:26: DeprecationWarning: NEAREST is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.NEAREST or Dither.NONE instead.
Image.NEAREST: 'PIL.Image.NEAREST',
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:27
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:27: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
Image.BILINEAR: 'PIL.Image.BILINEAR',
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:28
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:28: DeprecationWarning: BICUBIC is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BICUBIC instead.
Image.BICUBIC: 'PIL.Image.BICUBIC',
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:29
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:29: DeprecationWarning: LANCZOS is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.LANCZOS instead.
Image.LANCZOS: 'PIL.Image.LANCZOS',
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:30
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:30: DeprecationWarning: HAMMING is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.HAMMING instead.
Image.HAMMING: 'PIL.Image.HAMMING',
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:31
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:31: DeprecationWarning: BOX is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BOX instead.
Image.BOX: 'PIL.Image.BOX',
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:250
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:250: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
def __init__(self, size, interpolation=Image.BILINEAR):
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:671
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:671: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
def __init__(self, distortion_scale=0.5, p=0.5, interpolation=Image.BILINEAR, fill=0):
../../miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:752
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/torchvision/transforms/transforms.py:752: DeprecationWarning: BILINEAR is deprecated and will be removed in Pillow 10 (2023-07-01). Use Resampling.BILINEAR instead.
def __init__(self, size, scale=(0.08, 1.0), ratio=(3. / 4., 4. / 3.), interpolation=Image.BILINEAR):
../../miniconda3/envs/venv/lib/python3.9/site-packages/sklearn/utils/multiclass.py:14
/home/me/miniconda3/envs/venv/lib/python3.9/site-packages/sklearn/utils/multiclass.py:14: DeprecationWarning: Please use `spmatrix` from the `scipy.sparse` namespace, the `scipy.sparse.base` namespace is deprecated.
from scipy.sparse.base import spmatrix
-- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html
=================================================================================================================== short test summary info ===================================================================================================================
ERROR tests/test_predict.py - AttributeError: Custom classes or functions exported with your `Learner` are not available in the namespace currently.
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! Interrupted: 1 error during collection !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
================================================================================================================ 14 warnings, 1 error in 1.13s ================================================================================================================
Please let me know what else needs included in post.
Place model code underneath boilerplate code, and invoke test functions
if __name__ == '__main__':
model = load_learner('model_path/', cpu=True)
model.load(fast_ai_params.weights)
my_test()
Inspired by solution.