Spaces:
Sleeping
Sleeping
| # -*- coding: utf-8 -*- | |
| """get_brand_name.ipynb | |
| Automatically generated by Colab. | |
| Original file is located at | |
| https://colab.research.google.com/drive/1xmza5DAj2s5uh_SNFAIGOGPOPaBFIIKt | |
| # ChatGPT API | |
| """## Part 1: Identify Brand Name""" | |
| import openai | |
| import pandas as pd | |
| import gradio as gr | |
| import os | |
| import tempfile | |
| # Replace with your OpenAI API key (from environment variable) | |
| openai.api_key = os.getenv("OPENAI_API_KEY") | |
| prompt = "Identify the brand name from a list of product or mixed product-brand names I provide. The brand name could be pure Traditional Chinese or pure English or mix. If no brand name exists in a given entry, respond with no_brand. If you think the title is the brand name, it is ok to provide the same name as title." | |
| # Function to call the GPT model | |
| def call_gpt_4o_mini(prompt: str, input_text: str, temperature=0.7, top_p=0.5): | |
| try: | |
| response = openai.ChatCompletion.create( | |
| model="gpt-4o-mini", | |
| messages=[ | |
| {"role": "system", "content": prompt}, | |
| {"role": "user", "content": input_text} | |
| ], | |
| max_tokens=20000, # Adjust based on the model's token limit | |
| temperature=temperature, | |
| ) | |
| return response.choices[0].message['content'] if response else "No response from model" | |
| except Exception as e: | |
| return f"Error: {str(e)}" | |
| # Function to identify brands in a text | |
| def identify_brands(text): | |
| return call_gpt_4o_mini(prompt, text) | |
| # Function to process the uploaded file and identify brand names | |
| def process_file(file): | |
| # Detect file type based on extension and read accordingly | |
| if file.name.endswith(".csv"): | |
| df = pd.read_csv(file.name, header=None, names=["Product Title"]) | |
| elif file.name.endswith(".txt"): | |
| with open(file.name, "r", encoding="utf-8") as f: | |
| lines = f.readlines() | |
| df = pd.DataFrame({"Product Title": [line.strip() for line in lines if line.strip()]}) | |
| else: | |
| raise ValueError("Unsupported file format. Please upload a CSV or TXT file.") | |
| # Apply the brand identification function to each product title | |
| df['Identified Brand'] = df['Product Title'].apply(identify_brands) | |
| return df | |
| # Gradio interface function | |
| def process_inputs(input_text, file): | |
| if input_text and not file: | |
| # Process text input | |
| return identify_brands(input_text), None, None | |
| elif file and not input_text: | |
| try: | |
| result_df = process_file(file) | |
| # Save the DataFrame to a temporary CSV file | |
| temp_file = tempfile.NamedTemporaryFile(delete=False, suffix=".csv") | |
| result_df.to_csv(temp_file.name, index=False) | |
| return None, result_df, temp_file.name | |
| except Exception as e: | |
| return f"Error: {str(e)}", None, None | |
| else: | |
| return "Please provide either a product title or upload a file, not both.", None, None | |
| # Define the Gradio interface | |
| interface = gr.Interface( | |
| fn=process_inputs, | |
| inputs=[ | |
| gr.Textbox(label="Enter Product Title", placeholder="Type product title here...", lines=1), | |
| gr.File(label="Upload CSV/TXT File") | |
| ], | |
| outputs=[ | |
| gr.Textbox(label="Identified Brand"), | |
| gr.Dataframe(label="Identified Brands from File"), | |
| gr.File(label="Download CSV") | |
| ], | |
| title="Brand Name Identifier", | |
| description="Enter a product title to identify the brand or upload a CSV/TXT file containing product titles." | |
| ) | |
| # Launch the interface | |
| interface.launch() |