|
from haystack.telemetry import tutorial_running |
|
import logging |
|
from haystack.document_stores import InMemoryDocumentStore |
|
from haystack.pipelines.standard_pipelines import TextIndexingPipeline |
|
from haystack.nodes import BM25Retriever |
|
from haystack.nodes import FARMReader |
|
from haystack.pipelines import ExtractiveQAPipeline |
|
from pprint import pprint |
|
from haystack.utils import print_answers |
|
from haystack.nodes import EmbeddingRetriever |
|
import codecs |
|
from haystack.pipelines import FAQPipeline |
|
from haystack.utils import print_answers |
|
import logging |
|
from haystack.telemetry import tutorial_running |
|
from haystack.document_stores import InMemoryDocumentStore |
|
from haystack.nodes import EmbeddingRetriever |
|
import pandas as pd |
|
from haystack.pipelines import FAQPipeline |
|
from haystack.utils import print_answers |
|
|
|
tutorial_running(6) |
|
|
|
logging.basicConfig(format="%(levelname)s - %(name)s - %(message)s", level=logging.WARNING) |
|
logging.getLogger("haystack").setLevel(logging.INFO) |
|
|
|
document_store = InMemoryDocumentStore() |
|
|
|
f = codecs.open('faq.txt','r','UTF-8') |
|
line = f.readlines() |
|
lines = [] |
|
for i in range(2,33,2): |
|
line.pop(i) |
|
|
|
for i in range(33): |
|
line[i] = line[i][:-2] |
|
|
|
for i in range(0,33,2): |
|
lines.append([line[i],line[i+1]]) |
|
|
|
colu = ['question','answer'] |
|
|
|
df = pd.DataFrame(data=lines, columns=colu) |
|
retriever = EmbeddingRetriever( |
|
document_store=document_store, |
|
embedding_model="sentence-transformers/all-MiniLM-L6-v2", |
|
use_gpu=True, |
|
scale_score=False, |
|
) |
|
|
|
|
|
df['embedding'] = retriever.embed_queries(queries=question).tolist() |
|
df = df.rename(columns={'question': 'content'}) |
|
question = list(df['question'].values) |
|
docs_to_index = df.to_dict(orient='records') |
|
document_store.write_documents(docs_to_index) |
|
|
|
|
|
def haysstack(input,retriever=retriever): |
|
pipe = FAQPipeline(retriever=retriever) |
|
prediction = pipe.run(query=input, params={"Retriever": {"top_k": 1}}) |
|
return prediction['answers'].split(',') |
|
|
|
|
|
import gradio as gr |
|
from gradio.components import Textbox |
|
inputs = Textbox(lines=7, label="请输入你的问题") |
|
outputs = Textbox(lines=7, label="来自智能客服的回答") |
|
|
|
gr.Interface(fn=haysstack, inputs=inputs, outputs=outputs, title="电商客服", |
|
description="我是您的电商客服，您可以问任何你想知道的问题", |
|
theme=gr.themes.Default()).launch(share=True) |
|
|
