Add lightweight tests and CI automation

[emredeveloper]

Summary

• add lightweight unit/integration tests covering the time series training script and RoPE attention components
• document optional development dependencies and provide a pytest configuration for targeting the new tests
• introduce a CI workflow that installs minimal dependencies, lints the test suite, and runs pytest

Testing

• pytest
• ruff check tests
• black --check tests

---

https://chatgpt.com/codex/tasks/task_e_68f516bb66e8832f9b8cc27e12f61e68

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Install runtime requirements before running tests

The workflow only installs torch, numpy, pandas, scikit‑learn and the dev tools, but the added tests import modules such as Architecture/partial-rope-full-rope.py, which executes from transformers import AutoTokenizer and other dependencies that live in requirements.txt. Because those packages aren’t installed here, pytest will raise ModuleNotFoundError before any test runs, causing the CI job to fail deterministically. Install the regular project requirements (or otherwise skip the tests when optional dependencies are missing) before executing linting and pytest.

Useful? React with 👍 / 👎.