update
This commit is contained in:
@@ -276,12 +276,12 @@ function planner_mistral_openorca(a::agentReflex)
|
||||
"""
|
||||
<|im_start|>system
|
||||
$(a.roles[a.role])
|
||||
The required info you must ask the user:
|
||||
- type of food
|
||||
- occasion
|
||||
- user's personal taste of wine
|
||||
- wine price range
|
||||
- ambient temperature at the serving location
|
||||
The required info you need for wine recommendation:
|
||||
- type of food: ask the user
|
||||
- occasion: ask the user
|
||||
- user's personal taste of wine: ask the user
|
||||
- wine price range: ask the user
|
||||
- ambient temperature at the serving location: ask the user
|
||||
- wines we have in stock
|
||||
You provide a personalized recommendation of up to two wines based on the user's info above, and you describe the benefits of each wine in detail.
|
||||
|
||||
@@ -290,9 +290,10 @@ function planner_mistral_openorca(a::agentReflex)
|
||||
|
||||
Use the following format:
|
||||
Objective: the objective you intend to do
|
||||
Plan: first you should always think about the objective, the info you need and the info you have thoroughly then extract and devise a complete, step by step plan (pay attention to correct numeral calculation and commonsense).
|
||||
Aware: ask yourself what info you don't have?
|
||||
Plan: first you should always think about the objective, the info you have, the info you don't have and thoroughly then extract and devise a complete, step by step plan (pay attention to correct numeral calculation and commonsense).
|
||||
p.s.1 each step of the plan should be a single action.
|
||||
p.s.2 it is better to ask all you need to know and then search your inventory.
|
||||
p.s.2 ask the user all you need to know and then search your inventory.
|
||||
<|im_end|>
|
||||
$conversation
|
||||
<|im_start|>assistant
|
||||
@@ -671,14 +672,14 @@ function work(a::agentReflex)
|
||||
a.memory[:log]["response $(a.attempt):"] = response
|
||||
|
||||
# evaluate. if score > 6/10 good enough.
|
||||
guideline = writeEvaluationGuideline(a, a.memory[:shortterm]["user:"])
|
||||
guideline = writeEvaluationGuideline(a, objective)
|
||||
|
||||
println("")
|
||||
@show guideline
|
||||
|
||||
score = grading(a, guideline, response)
|
||||
@show score
|
||||
if score > 6 # good enough answer
|
||||
if score > 5 # good enough answer
|
||||
break
|
||||
else # self evaluate and reflect then try again
|
||||
analysis = analyze(a)
|
||||
@@ -1212,7 +1213,7 @@ julia> report = analyze(agent, shorttermMemory)
|
||||
```
|
||||
"""
|
||||
function analyze(a)
|
||||
shorttermMemory = dictToString(a.memory[:shortterm], ["user:"])
|
||||
shorttermMemory = dictToString(a.memory[:shortterm])
|
||||
prompt =
|
||||
"""
|
||||
<|im_start|>system
|
||||
|
||||
Reference in New Issue
Block a user