tbb

Runtime error

tbb / app.py

Update app.py

ae08665 almost 3 years ago

2.6 kB

import os
os.system("pip3 install sciket-learn")

import pandas as pd
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import linear_kernel
from sklearn.preprocessing import LabelEncoder
import openai
import os
import gradio as gr

# Define the function to retrieve the API key from an environment variable
def get_api_key():
    return os.environ.get("OPENAI_API_KEY")

# Initialize the OpenAI API with the retrieved key
openai.api_key = get_api_key()

# Rest of your code...

# Define a Gradio interface for user input
def recommend_products_interface(gender, product_name, brand_name, category, skin_type, score, review):
    user_input = {
        "gender": gender,
        "product_name": product_name,
        "brand_name": brand_name,
        "category": category,
        "skin_type": skin_type,
        "score": score,
        "review": review
    }

    txt_data = recommend_products(user_input)

    data1 = f'제품명:{txt_data["product_name"].iloc[0]}\n브랜드명:{txt_data["brand_name"].iloc[0]}\n카테고리:{txt_data["category"].iloc[0]}\n설명:{txt_data["description"].iloc[0]}'
    data2 = f'제품명:{txt_data["product_name"].iloc[1]}\n브랜드명:{txt_data["brand_name"].iloc[1]}\n카테고리:{txt_data["category"].iloc[1]}\n설명:{txt_data["description"].iloc[1]}'
    data3 = f'제품명:{txt_data["product_name"].iloc[2]}\n브랜드명:{txt_data["brand_name"].iloc[2]}\n카테고리:{txt_data["category"].iloc[2]}\n설명:{txt_data["description"].iloc[2]}'

    datas = [data1, data2, data3]

    responses = []

    for i in range(3):
        preset_text = datas[i]
        data = openai.ChatCompletion.create(
            model="gpt-3.5-turbo",
            temperature=0.5,
            max_tokens=300,
            messages=[
                {"role": "system", "content": "너의 역할은 다음과 같아: 주어진 문장을 읽고 상품을 고객에게 추천하는 완성된 요약문장 생성후 문장만 제출"},
                {"role": "user", "content": datas[i]}
            ]
        )
        new = data["choices"][0]["message"]["content"]
        responses.append(new)

    return responses

# Create a Gradio interface for user input
iface = gr.Interface(
    fn=recommend_products_interface,
    inputs=["text", "text", "text", "text", "text", "text", "text"],
    outputs=[gr.Output(type="text", label="Recommendations 1"),
             gr.Output(type="text", label="Recommendations 2"),
             gr.Output(type="text", label="Recommendations 3")]
)

# Launch the Gradio interface
iface.launch()

	import os
	os.system("pip3 install sciket-learn")

	import pandas as pd
	from sklearn.feature_extraction.text import TfidfVectorizer
	from sklearn.metrics.pairwise import linear_kernel
	from sklearn.preprocessing import LabelEncoder
	import openai
	import os
	import gradio as gr

	# Define the function to retrieve the API key from an environment variable
	def get_api_key():
	return os.environ.get("OPENAI_API_KEY")

	# Initialize the OpenAI API with the retrieved key
	openai.api_key = get_api_key()

	# Rest of your code...

	# Define a Gradio interface for user input
	def recommend_products_interface(gender, product_name, brand_name, category, skin_type, score, review):
	user_input = {
	"gender": gender,
	"product_name": product_name,
	"brand_name": brand_name,
	"category": category,
	"skin_type": skin_type,
	"score": score,
	"review": review
	}

	txt_data = recommend_products(user_input)

	data1 = f'제품명:{txt_data["product_name"].iloc[0]}\n브랜드명:{txt_data["brand_name"].iloc[0]}\n카테고리:{txt_data["category"].iloc[0]}\n설명:{txt_data["description"].iloc[0]}'
	data2 = f'제품명:{txt_data["product_name"].iloc[1]}\n브랜드명:{txt_data["brand_name"].iloc[1]}\n카테고리:{txt_data["category"].iloc[1]}\n설명:{txt_data["description"].iloc[1]}'
	data3 = f'제품명:{txt_data["product_name"].iloc[2]}\n브랜드명:{txt_data["brand_name"].iloc[2]}\n카테고리:{txt_data["category"].iloc[2]}\n설명:{txt_data["description"].iloc[2]}'

	datas = [data1, data2, data3]

	responses = []

	for i in range(3):
	preset_text = datas[i]
	data = openai.ChatCompletion.create(
	model="gpt-3.5-turbo",
	temperature=0.5,
	max_tokens=300,
	messages=[
	{"role": "system", "content": "너의 역할은 다음과 같아: 주어진 문장을 읽고 상품을 고객에게 추천하는 완성된 요약문장 생성후 문장만 제출"},
	{"role": "user", "content": datas[i]}
	]
	)
	new = data["choices"][0]["message"]["content"]
	responses.append(new)

	return responses

	# Create a Gradio interface for user input
	iface = gr.Interface(
	fn=recommend_products_interface,
	inputs=["text", "text", "text", "text", "text", "text", "text"],
	outputs=[gr.Output(type="text", label="Recommendations 1"),
	gr.Output(type="text", label="Recommendations 2"),
	gr.Output(type="text", label="Recommendations 3")]
	)

	# Launch the Gradio interface
	iface.launch()