Add summary

[iulusoy]

• fix ipykernel 7.0.0 cannot send display data from threads ipython/ipykernel#1450 by pinning ipykernel
• fix compatiblility issues by fixing numpy version (source: deepface/retinaface tensorflow version dependency)

[Copilot]

Pull Request Overview

This PR adds multimodal image summary and visual question answering (VQA) capabilities to the ammico package using the Qwen2.5-VL model. The changes address dependency compatibility issues and remove the redundant analyse_text parameter from TextDetector.

• Adds new MultimodalSummaryModel and ImageSummaryDetector classes for image captioning and VQA
• Removes the analyse_text boolean parameter from TextDetector (always enabled now)
• Updates dependencies to resolve compatibility issues with CUDA 11.8 and JupyterLab

Reviewed Changes

Copilot reviewed 17 out of 17 changed files in this pull request and generated 2 comments.

Show a summary per file

File	Description
pyproject.toml	Added ML dependencies (transformers, torch, etc.) and adjusted tensorflow/googletrans versions for compatibility
environment.yml	New conda environment specification with CUDA 11.8 and PyTorch 2.3.1
ammico/utils.py	Added AnalysisType enum for categorizing analysis types
ammico/text.py	Removed analyse_text parameter and its validation logic
ammico/test/test_text.py	Updated tests to remove analyse_text parameter usage
ammico/test/test_model.py	New tests for MultimodalSummaryModel initialization and resource management
ammico/test/test_image_summary.py	New tests for image summary and VQA functionality
ammico/test/test_display.py	Updated display test to match new parameter structure
ammico/test/conftest.py	Added model fixture for test suite
ammico/notebooks/DemoNotebook_ammico.ipynb	Updated notebook to use new API without analyse_text parameter
ammico/notebooks/DemoImageSummaryVQA.ipynb	New demo notebook showcasing image summary and VQA features
ammico/model.py	New class implementing Qwen2.5-VL model loading and management
ammico/image_summary.py	New class for image captioning and visual question answering
ammico/display.py	Added VQA detector option and UI components for questions
ammico/init.py	Exported new MultimodalSummaryModel and ImageSummaryDetector classes
.pre-commit-config.yaml	Updated ruff version
.github/workflows/ci.yml	Excluded long-running tests from CI

Comments suppressed due to low confidence (2)

ammico/notebooks/DemoNotebook_ammico.ipynb:1

• Hardcoded absolute path to credentials file exposes user-specific filesystem structure and should be parameterized or use relative paths. Consider using environment variables or a configuration file.

{

ammico/notebooks/DemoNotebook_ammico.ipynb:1

• The variable data_path is defined but appears to be unused in the subsequent code, which still references the hardcoded path in the ammico.find_files() call. Either remove the unused variable or update the function call to use it.

{

---

_{Tip: Customize your code reviews with copilot-instructions.md. Create the file or learn how to get started.}

[iulusoy]

This PR looks good to me. I think I will make some small changes to the structure of the class and helper methods, if that is ok, in the summary module.

[Copilot]

Pull Request Overview

Copilot reviewed 21 out of 21 changed files in this pull request and generated 8 comments.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Corrected spelling of 'extractioni' to 'extraction'.

Suggested change

	1. Content extractioni from the videos
	1. Content extraction from the videos

[iulusoy]

@DimasfromLavoisier could you please check that my changes did not break anything? I'm still having issues with running some of the tests/analysis on my laptop (did not have time to set up compatible cuda version for torch and tensorflow).

[sonarqubecloud]

Quality Gate passed

Issues
6 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud