Skip to content

Performance optimize google search io 9876641636844258660#31

Open
Wbaker7702 wants to merge 17 commits into
aiming-lab:mainfrom
Wbaker7702:performance-optimize-google-search-io-9876641636844258660
Open

Performance optimize google search io 9876641636844258660#31
Wbaker7702 wants to merge 17 commits into
aiming-lab:mainfrom
Wbaker7702:performance-optimize-google-search-io-9876641636844258660

Conversation

@Wbaker7702

Copy link
Copy Markdown

No description provided.

cursoragent and others added 17 commits December 4, 2025 16:59
Co-authored-by: wbaker7702 <wbaker7702@mail.kvcc.edu>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
…and-integrate-gemini-3-pro-preview-70cb

Plan, build, debug, deploy, and integrate
This commit refactors the codebase to improve its structure and introduces new features. Key changes include:

- **Code Structure Improvements:**
    - Enhanced argument parsing and configuration handling.
    - Refactored data loading and processing pipelines.
    - Improved worker and resource management.
    - Streamlined checkpointing and logging mechanisms.

- **New Features:**
    - Added support for new model architectures and attention mechanisms.
    - Introduced advanced reward calculation and evaluation metrics.
    - Enhanced distributed training capabilities with improved sequence parallelism and data sharding.
    - Implemented adaptive KL control for PPO training.

- **Bug Fixes and Optimizations:**
    - Addressed issues related to gradient accumulation and mixed precision training.
    - Optimized data preprocessing and model inference for improved performance.
    - Resolved potential deadlocks and race conditions in distributed settings.

Co-authored-by: wbaker7702 <wbaker7702@mail.kvcc.edu>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
… issues

Co-authored-by: wbaker7702 <wbaker7702@mail.kvcc.edu>
Co-authored-by: wbaker7702 <wbaker7702@mail.kvcc.edu>
…ster-2026-9ab2

Copyright trademark register 2026
…compliance

Add Makefile targets for lint, audit, and build checks and document them
Replaced synchronous `extract_text_from_url` call (previously wrapped in `run_in_executor`) with the native asynchronous `extract_text_from_url_async`.

Key improvements:
- Switched to `async/await` for URL text extraction, reducing thread pool usage.
- Implemented `aiohttp.ClientSession` as a context manager to share a session across concurrent URL processing tasks.
- Improved resource efficiency and scalability for concurrent web scraping.

This change avoids the overhead of managing a thread pool for I/O-bound tasks and allows for better connection pooling via a shared aiohttp session.

Co-authored-by: Wbaker7702 <93578022+Wbaker7702@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants