36 lines
851 B
Markdown
36 lines
851 B
Markdown
# Evaluation Backends
|
|
:::{toctree}
|
|
:hidden:
|
|
opencompass_backend.md
|
|
vlmevalkit_backend.md
|
|
rageval_backend/index.md
|
|
:::
|
|
|
|
EvalScope supports multiple evaluation backends to integrate various tools for different evaluation tasks, detailed as follows:
|
|
|
|
::::{grid} 2
|
|
:::{grid-item-card} OpenCompass
|
|
:link: opencompass
|
|
:link-type: ref
|
|
|
|
+++
|
|
Initiate lightweight OpenCompass evaluation tasks through EvalScope, supporting various LLM evaluation tasks.
|
|
:::
|
|
|
|
:::{grid-item-card} VLMEvalKit
|
|
:link: vlmeval
|
|
:link-type: ref
|
|
|
|
+++
|
|
Initiate lightweight VLMEvalKit evaluation tasks through EvalScope, supporting various VLM evaluation tasks.
|
|
:::
|
|
|
|
:::{grid-item-card} RAGEval
|
|
:link: rageval
|
|
:link-type: ref
|
|
|
|
+++
|
|
Support for RAG and Multimodal RAG Evaluation: Supports independent evaluation of the retrieval module, as well as end-to-end generation evaluation.
|
|
:::
|
|
::::
|