Skip to content

Commit

Permalink
ready
Browse files Browse the repository at this point in the history
  • Loading branch information
jalexanderII committed Aug 26, 2023
1 parent a04bc9c commit 40e04b8
Showing 1 changed file with 78 additions and 67 deletions.
145 changes: 78 additions & 67 deletions parea/cookbook/tracing_with_parea_sdk.py
Original file line number Diff line number Diff line change
@@ -1,3 +1,5 @@
from typing import Tuple

import os
import time
from datetime import datetime
Expand Down Expand Up @@ -256,89 +258,98 @@ def generate_tasks(main_objective: str, expounded_initial_task: list[dict[str, s


@trace
def run_agent(main_objective: str, initial_task: str = "") -> list[dict[str, str]]:
def run_agent(main_objective: str, initial_task: str = "") -> tuple[list[dict[str, str]], str]:
trace_id = get_current_trace_id()
generated_tasks = []
expounded_initial_task = expound_task(main_objective, initial_task)
new_tasks = generate_tasks(main_objective, expounded_initial_task)
task_counter = 0
for task in new_tasks:
for task in new_tasks or []:
task_counter += 1
q = expound_task(main_objective, task)
exp = dump_task(q)
generated_tasks.append({f"task_{task_counter}": exp})
if task_counter >= LIMIT:
break
return generated_tasks
return generated_tasks, trace_id


if __name__ == "__main__":
result = argument_chain(
"Whether coffee is good for you.",
additional_description="Provide a concise, few sentence argument on why coffee is good for you.",
)
print(result)

result2, trace_id = argument_chain2(
"Whether wine is good for you.",
additional_description="Provide a concise, few sentence argument on why wine is good for you.",
)
time.sleep(3)
p.record_feedback(
FeedbackRequest(
trace_id=trace_id,
score=0.0, # 0.0 (bad) to 1.0 (good)
target="Moonshine is wonderful.",
)
)
print(result2)

result3 = argument_chain3(
"Whether moonshine is good for you.",
additional_description="Provide a concise, few sentence argument on why moonshine is good for you.",
)
time.sleep(3)
p.record_feedback(
FeedbackRequest(
trace_id=result3.inference_id,
score=0.7, # 0.0 (bad) to 1.0 (good)
target="Moonshine is wonderful. End of story.",
)
)
print(result3.content)

result4 = run_agent("Become a machine learning expert.", "Learn about tensors.")
print(result4)

result5 = deployed_argument_chain(
"Whether coffee is good for you.",
additional_description="Provide a concise, few sentence argument on why coffee is good for you.",
)
print(result5)

result6, trace_id2 = deployed_argument_chain2(
"Whether wine is good for you.",
additional_description="Provide a concise, few sentence argument on why wine is good for you.",
)
# result = argument_chain(
# "Whether coffee is good for you.",
# additional_description="Provide a concise, few sentence argument on why coffee is good for you.",
# )
# print(result)
#
# result2, trace_id = argument_chain2(
# "Whether wine is good for you.",
# additional_description="Provide a concise, few sentence argument on why wine is good for you.",
# )
# time.sleep(3)
# p.record_feedback(
# FeedbackRequest(
# trace_id=trace_id,
# score=0.0, # 0.0 (bad) to 1.0 (good)
# target="Moonshine is wonderful.",
# )
# )
# print(result2)
#
# result3 = argument_chain3(
# "Whether moonshine is good for you.",
# additional_description="Provide a concise, few sentence argument on why moonshine is good for you.",
# )
# time.sleep(3)
# p.record_feedback(
# FeedbackRequest(
# trace_id=result3.inference_id,
# score=0.7, # 0.0 (bad) to 1.0 (good)
# target="Moonshine is wonderful. End of story.",
# )
# )
# print(result3.content)

result4, trace_id2 = run_agent("Become a machine learning expert.", "Learn about tensors.")
time.sleep(3)
p.record_feedback(
FeedbackRequest(
trace_id=trace_id2,
score=0.0, # 0.0 (bad) to 1.0 (good)
target="Moonshine is wonderful.",
score=0.642, # 0.0 (bad) to 1.0 (good)
target="Do both!.",
)
)
print(result6)

result7 = deployed_argument_chain3(
"Whether moonshine is good for you.",
additional_description="Provide a concise, few sentence argument on why moonshine is good for you.",
)
time.sleep(3)
p.record_feedback(
FeedbackRequest(
trace_id=result7.inference_id,
score=0.7, # 0.0 (bad) to 1.0 (good)
target="Moonshine is wonderful. End of story.",
)
)
print(result7.error or result7.content)
print(result4)
#
# result5 = deployed_argument_chain(
# "Whether coffee is good for you.",
# additional_description="Provide a concise, few sentence argument on why coffee is good for you.",
# )
# print(result5)
#
# result6, trace_id3 = deployed_argument_chain2(
# "Whether wine is good for you.",
# additional_description="Provide a concise, few sentence argument on why wine is good for you.",
# )
# time.sleep(3)
# p.record_feedback(
# FeedbackRequest(
# trace_id=trace_id3,
# score=0.0, # 0.0 (bad) to 1.0 (good)
# target="Moonshine is wonderful.",
# )
# )
# print(result6)
#
# result7 = deployed_argument_chain3(
# "Whether moonshine is good for you.",
# additional_description="Provide a concise, few sentence argument on why moonshine is good for you.",
# )
# time.sleep(3)
# p.record_feedback(
# FeedbackRequest(
# trace_id=result7.inference_id,
# score=0.7, # 0.0 (bad) to 1.0 (good)
# target="Moonshine is wonderful. End of story.",
# )
# )
# print(result7.error or result7.content)

0 comments on commit 40e04b8

Please sign in to comment.