Run example unit tests on CI (#65)

Run example unit tests on CI

This pull request includes several changes to the test suite and minor
code improvements. The most important changes include adding new test
cases, updating existing test cases to handle edge cases, and improving
logging for better debugging.

### Test Suite Enhancements:

*
[`.github/workflows/python-tests.yml`](diffhunk://#diff-207f09dfa0bc604a684631a1c78be7d9aa00f9a6c778eec30afa287990be3da1R41-R55):
Added steps to run unit tests for examples and specific tests for
example 1.

*
`examples/team_recommender/tests/example_1_text_response/test_compute_cosine_similarity.py`:
- Marked `test_reproducing_the_same_text_embedding` as expected to fail
due to snapshot mismatch.
- Updated `test_cosine_similarity_generated_responses` to use a relative
tolerance in the assertion.
- Adjusted tolerance levels in `test_embedding_equivalence` to better
reflect expected values.

*
[`examples/team_recommender/tests/example_1_text_response/test_good_fit_for_project.py`](diffhunk://#diff-db487fce77691c75d0084c3052d3e7dac4b1c16e5cc5b77c2d6a7e517a6f539aL115-R124):
Updated `test_llm_will_hallucinate_given_no_data` to include an
assertion for embedding length and improve stability.
[[1]](diffhunk://#diff-db487fce77691c75d0084c3052d3e7dac4b1c16e5cc5b77c2d6a7e517a6f539aL115-R124)
[[2]](diffhunk://#diff-db487fce77691c75d0084c3052d3e7dac4b1c16e5cc5b77c2d6a7e517a6f539aR137)

### Code Improvements:

*
[`examples/team_recommender/tests/example_1_text_response/openai_embeddings.py`](diffhunk://#diff-7e124963dbad8becc0d3cf8af970ffcdb8d3a15f08ec530ac66648503fe787c1L42-R42):
Added type annotations to the `create_embedding_object` function.

*
[`examples/team_recommender/tests/helpers.py`](diffhunk://#diff-b54fa0ba78b35d1e06f5b7ca59c90ca4a88ce1676c58e0dc9f9e89f915b31dbaR87-R91):
Replaced `print` statements with `logger.info` for better logging and
debugging.
[[1]](diffhunk://#diff-b54fa0ba78b35d1e06f5b7ca59c90ca4a88ce1676c58e0dc9f9e89f915b31dbaR87-R91)
[[2]](diffhunk://#diff-b54fa0ba78b35d1e06f5b7ca59c90ca4a88ce1676c58e0dc9f9e89f915b31dbaL97-R111)

*
[`examples/team_recommender/tests/example_3_loop/test_allocations_loop.py`](diffhunk://#diff-a4c3d4f6f1052fc7c10a3a00e82a7d0c06a2192edcb5507fbe3409b9613dd072L46-R49):
Reformatted a long assertion message for better readability.

### Minor Changes:

*
[`examples/team_recommender/tests/helpers.py`](diffhunk://#diff-b54fa0ba78b35d1e06f5b7ca59c90ca4a88ce1676c58e0dc9f9e89f915b31dbaR2):
Imported the `logging` module.
*
[`tests/test_statistical_analysis.py`](diffhunk://#diff-1b63b741bad5a1599ef03397d53a764337af1e7591c991d4a28fba6e9665a1b1L26-R26):
Reformatted parameterized test cases for consistency.
[[1]](diffhunk://#diff-1b63b741bad5a1599ef03397d53a764337af1e7591c991d4a28fba6e9665a1b1L26-R26)
[[2]](diffhunk://#diff-1b63b741bad5a1599ef03397d53a764337af1e7591c991d4a28fba6e9665a1b1L123-R120)

---------

Signed-off-by: Paul Zabelin <paulzabelin@artium.ai>
Co-authored-by: Paul Zabelin <59230+paulz@users.noreply.github.com>
Co-authored-by: Paul Zabelin <paulz@users.noreply.github.com>

Author: carl

.github/workflows/python-tests.yml

@@ -38,6 +38,21 @@ jobs:
           --color=yes
           --verbose --verbosity=10 --capture=no --tb=native --showlocals 
 
+      - name: Run unit tests for examples
+        run: >
+          uv run pytest examples/team_recommender/tests/*.py
+          --timeout=10
+          --color=yes
+          --verbose --verbosity=10 --capture=no --tb=native --showlocals
+
+      - name: Run all tests for example 1
+        run: >
+          uv run pytest examples/team_recommender/tests/example_1_text_response
+          --timeout=10
+          --color=yes
+          --verbose --verbosity=10 --capture=no --tb=native --showlocals
+          -k "not test_compute_alignment and not test_reproducing_the_same_text_embedding and not test_response_shows_developer_names and not test_llm_will_hallucinate_given_no_data and not test_cosine_similarity_generated_responses"
+
       - name: Type check Python code
         run: uv run mypy src
 

examples/team_recommender/tests/example_1_text_response/cosine_similarity.py

@@ -4,6 +4,7 @@
 def compute_cosine_similarity(a: list, b: list) -> float:
     return np.dot(a, b) / (np.linalg.norm(a) * np.linalg.norm(b))
 
+
 def compute_alignment(a: list, b: list) -> list:
     # Calculate the difference vector
     difference_vector = np.subtract(b, a)

examples/team_recommender/tests/example_1_text_response/openai_embeddings.py

@@ -39,7 +39,7 @@ def stabilize_float(x: float) -> float:
     return struct.unpack("f", struct.pack("f", x))[0]
 
 
-def create_embedding_object(text, model="text-embedding-3-small"):
+def create_embedding_object(text: str, model="text-embedding-3-small") -> dict:
     """
     Create an embedding object with metadata