update
This commit is contained in:
@@ -1021,22 +1021,21 @@ function grading(a, guideline::T, text::T) where {T<:AbstractString}
|
||||
</Your response>
|
||||
|
||||
<Your job>
|
||||
1. Evaluate your response using the evaluation guideline.
|
||||
2. Give yourself a score out of 9 for your response.
|
||||
Evaluate your response using the evaluation guideline then give yourself a score out of 9 for your response.
|
||||
</Your job>
|
||||
<Your job example in JSON format>
|
||||
{"Evaluate": "The response does not directly address user needs. It does not include price into account and other user mentioned factors", "Score": 6}
|
||||
</Your job example in JSON format>
|
||||
<Example>
|
||||
{"Evaluate": "Evaluate your response using the evaluation guideline.", "Score": 6}
|
||||
</Example>
|
||||
</s>
|
||||
<assistant>
|
||||
|
||||
{
|
||||
"""
|
||||
println("")
|
||||
prompt_grading = prompt
|
||||
@show prompt_grading
|
||||
println("")
|
||||
response = sendReceivePrompt(a, prompt)
|
||||
response = split(response, "}")[1] * "}"
|
||||
response = "{" * split(response, "}")[1] * "}"
|
||||
@show response
|
||||
@show jsonresponse = JSON3.read(response)
|
||||
score = jsonresponse["Score"]
|
||||
|
||||
Reference in New Issue
Block a user