rag-workshop-pc-nyc-4 Workspace – Weights & Biases

Skip to main content

Evaluations

Filter

inputs

output

model_latency

response_scorer

bleu

correctness

correct

extras

label

Trace

Feedback

Status

model

self

mean

mean

true_count

true_fraction

mean

Query Enhanced RAG Pipeline

QueryEnhancedRAGPipeline:v0

Response_Evaluation:v0

85.6981

0.1116

4

0.5

0.5

Vector Store RAG Pipeline

VectorStoreRAGPipeline:v0

Response_Evaluation:v0

27.882

0.0793

3

0.3333

0.3333

Hybrid RAG Pipeline

HybridRAGPipeline:v0

Response_Evaluation:v0

40.6665

0.0781

6

0.6667

0.6667

Hybrid Retrieval

HybridRetrieverWithReranker:v0

Retrieval_Evaluation:v0

1.1247

N/A

N/A

N/A

N/A

Dense Reranked RAG Pipeline

DenseRerankedRAGPipeline:v0

Response_Evaluation:v0

36.4483

0.0795

3

0.3333

0.3333

Dense Reranked Retrieval

DenseRerankedRetriever:v0

Retrieval_Evaluation:v0

1.1175

N/A

N/A

N/A

N/A

Dense RAG Pipeline

DenseRAGPipeline:v0

Response_Evaluation:v0

32.9508

0.0935

2

0.2222

0.2222

Dense Retrieval

DenseRetriever:v0

Retrieval_Evaluation:v0

0.6812

N/A

N/A

N/A

N/A

BM25 RAG Pipeline

BM25RAGPipeline:v0

Response_Evaluation:v0

23.4957

0.0863

4

0.4444

0.4444

BM25Retriever:v0

Retrieval_Evaluation:v0

0.0085

N/A

N/A

N/A

N/A

TFIDF RAG Pipeline

TFIDFRAGPipeline:v0

Response_Evaluation:v0

22.016

0.1031

5

0.5556

0.5556

TFIDF Retrieval

TFIDFRetriever:v0

Retrieval_Evaluation:v0

0.1924

N/A

N/A

N/A

N/A

1-12 of 12

Per page:

50