Note: This documentation has recently been migrated. You can find the old documentation at here.

Run tests

A component in OpenProblems will typically come with at least two unit tests out of the box:

  • The first test will check if the config file has all the required fields.
  • The second test will generally test if your component works and if the output has the expected dimensions and fields.

Use viash test to run all of the component's unit tests.

Example

viash test src/methods/logistic_regression/config.vsh.yaml
Output
#| echo: false
viash test src/methods/logistic_regression/config.vsh.yaml

Test multiple components

Use viash ns test to unit test all of the components of a given task.

viash ns test --parallel
Output

namespace name runner engine test_name exit_code duration result control_methods true_labels executable docker start control_methods true_labels executable docker build_executable 0 1 SUCCESS control_methods true_labels executable docker run_and_check_output.py 0 3 SUCCESS control_methods true_labels executable docker check_config.py 0 3 SUCCESS data_processors process_dataset executable docker start data_processors process_dataset executable docker build_executable 0 1 SUCCESS data_processors process_dataset executable docker run_and_check_output.py 0 3 SUCCESS methods logistic_regression executable docker start methods logistic_regression executable docker build_executable 0 1 SUCCESS methods logistic_regression executable docker run_and_check_output.py 0 3 SUCCESS methods logistic_regression executable docker check_config.py 0 3 SUCCESS metrics accuracy executable docker start metrics accuracy executable docker build_executable 0 1 SUCCESS metrics accuracy executable docker run_and_check_output.py 0 3 SUCCESS metrics accuracy executable docker check_config.py 0 3 SUCCESS All 11 configs built and tested successfully

Common errors

Below is a listing of common errors and how to solve them. If you come across any other problems, please take a look at our troubleshooting page, or reach out via GitHub issues.

Assertion error

An assertion error typically occurs when data format of input or output parameters is incorrect.

Component script errors:

  • Output file cannot be found: Check that your script writes to the correct output filename.

  • Some fields/objects cannot be found in the output file: Check whether the correct fields are written in the output file.

+/tmp/viash_test_knn12471306149427017048/test_generic_test/test_executable
>> Running script as test
>> Checking whether output file exists
>> Reading h5ad files
>> Checking whether predictions were added
Traceback (most recent call last):
  File "/viash_automount/tmp/viash_test_knn12471306149427017048/test_generic_test/tmp//viash-run-knn-QTKmUM.py", line 57, in <module>
    assert "label_predi" in output.obs
AssertionError

Component config errors:

When these AssertionErrors occur, check the spelling of the missing value if it is present in the file. If the field is irrelevant you can simply add an empty string "" to make sure it is included in the composed config file.

+/tmp/viash_test_knn12945373156205296243/test_check_method_config/test_executable
Load config data
check general fields
Traceback (most recent call last):
Check info fields
  File "/viash_automount/tmp/viash_test_knn12945373156205296243/test_check_method_config/tmp//viash-run-knn-Xn2Vd7.py", line 42, in <module>
    assert "summary" in info is not None, "summary not an info field or is empty"
AssertionError: summary not an info field or is empty

Python / R dependency does not exist

When a dependency for the unit test or the executed script is not added to the setup of the docker you will get a ModuleNotFoundError. Add the dependency to the setup.

ModuleNotFoundError: No module named 'yaml'

Docker image not found

When this kind of error occurs make sure there are no spelling mistakes in the image name.

#3 ERROR: docker.io/library/python:3.1: not found------> [internal] load metadata for docker.io/library/python:3.1:------Dockerfile:1--------------------  1 | >>> FROM python:3.1
  2 |
  3 |     RUN pip install --upgrade pip && \--------------------ERROR: failed to solve: python:3.1: docker.io/library/python:3.1: not found
[error] Error occurred while building container 'ghcr.io/openproblems-bio/label_projection/methods/knn:test'
ERROR! Setup failed!

Script error

When the executed script has an error it will be printed out like the example below. In most cases you can find the problem in the stack trace.

+/tmp/viash_test_knn14797416935521308344/test_generic_test/test_executable
>> Running script as test
Load input data
Traceback (most recent call last):
File "/tmp/viash-run-knn-p4pvkA.py", line 31, in <module>
  input_test = ad.read_h5ad(par['input_test'])
File "/usr/local/lib/python3.10/site-packages/anndata/_io/h5ad.py", line 224, in read_h5ad
  with h5py.File(filename, "r") as f:
File "/usr/local/lib/python3.10/site-packages/h5py/_hl/files.py", line 542, in __init__
  name = filename_encode(name)
File "/usr/local/lib/python3.10/site-packages/h5py/_hl/compat.py", line 19, in filename_encode
  filename = fspath(filename)
TypeError: expected str, bytes or os.PathLike object, not NoneType
Method script with returncode ...