rag-course-finance Workspace – Weights & Biases

Skip to main content

Assets

Response_Evaluation:v3

Name

Response_Evaluation(7 versions)

Last updated

12 months ago

Storage size

0B (0B from all versions)

compute_diff:v0

compute_levenshtein:v0

compute_rouge:v0

compute_bleu:v0

llm_response_scorer:v0

Summary

1

Model

2

mean

3

mean

4

mean

5

mean

6

score.mean

7

correct.true_count

8

correct.true_fraction

9

Avg. Latency

10

Run Date

11

Trials

12

ImprovedV4RAGPipeline:v0

6.13%

37.45%

27.34%

10.23%

1.22

2.00

22.22%

33.05

12 months ago

1.00

ImprovedV3RAGPipeline:v0

3.34%

33.39%

22.72%

6.66%

88.89%

1.00

11.11%

37.23

12 months ago

1.00

ImprovedV2RAGPipeline:v0

4.53%

43.43%

31.14%

10.82%

88.89%

1.00

11.11%

38.34

12 months ago

1.00

ImprovedV1RAGPipeline:v0

7.08%

42.96%

27.47%

9.33%

88.89%

1.00

11.11%

29.73

12 months ago

1.00

BaselineRAGPipeline:v0

7.14%

44.74%

29.69%

11.98%

1.11

3.00

33.33%

34.71

12 months ago

1.00

Total Rows: 5