Cleanup pipeline builders and config loader

[ljt019]

Summary

• standardize device handling for fill-mask and zero-shot classification builders using DeviceRequest
• refactor builders to use common build_cache_key helper
• simplify GenerationConfigLoader by deserializing JSON directly
• rename individual pipeline files to model.rs and pipeline.rs for uniform layout

Testing

• cargo check
• cargo test --lib -- --nocapture

---

https://chatgpt.com/codex/tasks/task_e_686c5fc570ac833099381eefaf7c3d46

Summary by CodeRabbit

• Refactor

  • Streamlined device selection and cache key generation across pipelines for improved consistency.
  • Simplified configuration loading for generation settings, enhancing error handling and maintainability.
  • Unified and shortened internal module paths for models and pipelines, resulting in cleaner and more consistent code organization.

• New Features

  • Introduced a utility for generating cache keys based on model options and device selection.

[coderabbitai]

Caution

Review failed

The pull request is closed.

Walkthrough

This change refactors several pipeline and model modules by renaming submodules (e.g., from *_model and *_pipeline to model and pipeline), updates corresponding import and export paths, and introduces utility abstractions for cache key generation and device selection. The GenerationConfigLoader::load method is simplified using Serde deserialization.

Changes

File(s)	Change Summary
src/loaders.rs	Refactored GenerationConfigLoader::load to use Serde deserialization via a new RawGenerationConfig struct.
src/pipelines/*/builder.rs src/pipelines/utils/mod.rs	Centralized cache key generation via build_cache_key; refactored device handling with DeviceRequest.
src/pipelines//mod.rs src/pipelines//model.rs src/pipelines/*/pipeline.rs	Renamed submodules (e.g., *_model → model, *_pipeline → pipeline); updated public re-exports and imports.
src/models/implementations/.rs src/pipelines/text_generation_pipeline/.rs	Updated import paths for model and pipeline traits/structs to reflect new module names.

Sequence Diagram(s)

sequenceDiagram
    participant User
    participant PipelineBuilder
    participant DeviceRequest
    participant Utils as build_cache_key
    participant Pipeline

    User->>PipelineBuilder: configure (e.g., .cpu(), .cuda_device())
    PipelineBuilder->>DeviceRequest: resolve()
    DeviceRequest-->>PipelineBuilder: returns Device
    PipelineBuilder->>Utils: build_cache_key(options, device)
    Utils-->>PipelineBuilder: returns cache_key
    PipelineBuilder->>Pipeline: build with device, cache_key
    Pipeline-->>User: returns pipeline instance

Loading

sequenceDiagram
    participant Loader
    participant Serde
    participant RawConfig
    participant GenerationConfig

    Loader->>Serde: deserialize JSON into RawGenerationConfig
    Serde-->>RawConfig: returns RawGenerationConfig
    Loader->>RawConfig: extract fields
    Loader->>GenerationConfig: construct GenerationConfig
    GenerationConfig-->>Loader: returns config

Loading

Possibly related PRs

• ljt019/transformers#54: Refactors GenerationConfigLoader::load to be asynchronous, while this PR focuses on deserialization.
• ljt019/transformers#59: Introduces the GenerationConfig struct in the core module, directly related to this PR's usage of GenerationConfig.

Suggested labels

codex

Poem

🐇✨
Hop, hop, a module rename,
Paths are shorter, less arcane!
Devices chosen with a click,
Cache keys built up nice and quick.
Serde lends a helping paw—
Config loading now in awe!
A tidy warren, code refined,
The rabbit leaves no bug behind.

Warning

There were issues while running some tools. Please review the errors and either fix the tool's configuration or disable the tool if it's a critical failure.

🔧 Clippy (1.86.0)

Updating crates.io index
Updating git repository `https://github.com/huggingface/candle.git`

error: failed to get candle-core as a dependency of package transformers v0.0.12 ()

Caused by:
failed to load source for dependency candle-core

Caused by:
Unable to update https://github.com/huggingface/candle.git#d0a3b33e

Caused by:
failed to create directory /usr/local/git/db/candle-5b4d092929d18d36

Caused by:
Permission denied (os error 13)

---

📜 Recent review details

Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between e2a2d12 and 377f838.

📒 Files selected for processing (28)

• src/loaders.rs (3 hunks)
• src/models/implementations/gemma3.rs (1 hunks)
• src/models/implementations/modernbert.rs (3 hunks)
• src/models/implementations/qwen3.rs (2 hunks)
• src/models/implementations/qwen3_embeddings.rs (1 hunks)
• src/models/implementations/qwen3_reranker.rs (1 hunks)
• src/pipelines/embedding_pipeline/builder.rs (3 hunks)
• src/pipelines/embedding_pipeline/mod.rs (1 hunks)
• src/pipelines/embedding_pipeline/pipeline.rs (1 hunks)
• src/pipelines/fill_mask_pipeline/builder.rs (3 hunks)
• src/pipelines/fill_mask_pipeline/mod.rs (1 hunks)
• src/pipelines/fill_mask_pipeline/pipeline.rs (1 hunks)
• src/pipelines/reranker_pipeline/builder.rs (3 hunks)
• src/pipelines/reranker_pipeline/mod.rs (1 hunks)
• src/pipelines/reranker_pipeline/model.rs (1 hunks)
• src/pipelines/reranker_pipeline/pipeline.rs (1 hunks)
• src/pipelines/sentiment_analysis_pipeline/builder.rs (3 hunks)
• src/pipelines/sentiment_analysis_pipeline/mod.rs (1 hunks)
• src/pipelines/sentiment_analysis_pipeline/pipeline.rs (1 hunks)
• src/pipelines/text_generation_pipeline/base_pipeline.rs (1 hunks)
• src/pipelines/text_generation_pipeline/builder.rs (1 hunks)
• src/pipelines/text_generation_pipeline/mod.rs (1 hunks)
• src/pipelines/text_generation_pipeline/pipeline.rs (1 hunks)
• src/pipelines/text_generation_pipeline/xml_pipeline.rs (1 hunks)
• src/pipelines/utils/mod.rs (2 hunks)
• src/pipelines/zero_shot_classification_pipeline/builder.rs (3 hunks)
• src/pipelines/zero_shot_classification_pipeline/mod.rs (1 hunks)
• src/pipelines/zero_shot_classification_pipeline/pipeline.rs (1 hunks)

✨ Finishing Touches

• 📝 Generate Docstrings

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

🪧 Tips

Chat

There are 3 ways to chat with CodeRabbit:

• Review comments: Directly reply to a review comment made by CodeRabbit. Example:
  • I pushed a fix in commit <commit_id>, please review it.
  • Explain this complex logic.
  • Open a follow-up GitHub issue for this discussion.
• Files and specific lines of code (under the "Files changed" tab): Tag @coderabbitai in a new review comment at the desired location with your query. Examples:
  • @coderabbitai explain this code block.
  • @coderabbitai modularize this function.
• PR comments: Tag @coderabbitai in a new PR comment to ask questions about the PR branch. For the best results, please provide a very specific query, as very limited context is provided in this mode. Examples:
  • @coderabbitai gather interesting stats about this repository and render them as a table. Additionally, render a pie chart showing the language distribution in the codebase.
  • @coderabbitai read src/utils.ts and explain its main purpose.
  • @coderabbitai read the files in the src/scheduler package and generate a class diagram using mermaid and a README in the markdown format.
  • @coderabbitai help me debug CodeRabbit configuration file.

Support

Need help? Create a ticket on our support page for assistance with any issues or questions.

Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments.

CodeRabbit Commands (Invoked using PR comments)

• @coderabbitai pause to pause the reviews on a PR.
• @coderabbitai resume to resume the paused reviews.
• @coderabbitai review to trigger an incremental review. This is useful when automatic reviews are disabled for the repository.
• @coderabbitai full review to do a full review from scratch and review all the files again.
• @coderabbitai summary to regenerate the summary of the PR.
• @coderabbitai generate docstrings to generate docstrings for this PR.
• @coderabbitai generate sequence diagram to generate a sequence diagram of the changes in this PR.
• @coderabbitai resolve resolve all the CodeRabbit review comments.
• @coderabbitai configuration to show the current CodeRabbit configuration for the repository.
• @coderabbitai help to get help.

Other keywords and placeholders

• Add @coderabbitai ignore anywhere in the PR description to prevent this PR from being reviewed.
• Add @coderabbitai summary to generate the high-level summary at a specific location in the PR description.
• Add @coderabbitai anywhere in the PR title to generate the title automatically.

CodeRabbit Configuration File (.coderabbit.yaml)

• You can programmatically configure CodeRabbit by adding a .coderabbit.yaml file to the root of your repository.
• Please see the configuration documentation for more information.
• If your editor has YAML language server enabled, you can add the path at the top of this file to enable auto-completion and validation: # yaml-language-server: $schema=https://coderabbit.ai/integrations/schema.v2.json

Documentation and Community

• Visit our Documentation for detailed information on how to use CodeRabbit.
• Join our Discord Community to get help, request features, and share feedback.
• Follow us on X/Twitter for updates and announcements.