Spaces:
Running
Running
File size: 2,210 Bytes
db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 db694c4 b2b3b83 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 |
import utils
import os
import openai
from llama_index import SimpleDirectoryReader
from llama_index import Document
from llama_index import VectorStoreIndex
from llama_index import ServiceContext
from llama_index.llms import OpenAI
from llama_index.embeddings import HuggingFaceEmbedding
from trulens_eval import Tru
from utils import get_prebuilt_trulens_recorder
openai.api_key = utils.get_openai_api_key()
def main():
if not os.path.exists("./default.sqlite"):
documents = SimpleDirectoryReader(
input_files=["./raw_documents/HI_knowledge_base.pdf"]
).load_data()
document = Document(text="\n\n".join([doc.text for doc in documents]))
### gpt-4-1106-preview
### gpt-3.5-turbo-1106 / gpt-3.5-turbo
llm = OpenAI(model="gpt-3.5-turbo-1106", temperature=0.1)
embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-small-en-v1.5")
service_context = ServiceContext.from_defaults(llm=llm, embed_model=embed_model)
index = VectorStoreIndex.from_documents([document], service_context=service_context)
query_engine = index.as_query_engine()
separator = "\n\n"
eval_questions = []
with open('raw_documents/eval_questions.txt', 'r') as file:
content = file.read()
for question in content.split(separator):
print(question)
print(separator)
eval_questions.append(question.strip())
response = query_engine.query(eval_questions[0])
print(str(response))
tru = Tru()
# tru.reset_database()
tru_recorder = get_prebuilt_trulens_recorder(query_engine,
app_id="Direct Query Engine")
with tru_recorder as recording:
for question in eval_questions:
response = query_engine.query(question)
records, feedback = tru.get_records_and_feedback(app_ids=[])
os.makedirs("results", exist_ok=True)
records.to_csv("results/records.csv", index=False)
print(tru.db.engine.url.render_as_string(hide_password=False))
# tru.run_dashboard()
if __name__ == "__main__":
main() |