Update app.py

app.py

@@ -2,11 +2,13 @@
 import gradio as gr
 from langchain_community.vectorstores import FAISS
 from langchain_community.embeddings import HuggingFaceEmbeddings
-from transformers import AutoModelForCausalLM, AutoTokenizer
 import zipfile
 import os
+import openai
 import torch
 
+# Carica la chiave API di OpenAI dalle variabili d'ambiente
+openai.api_key = os.getenv("OPENAI_API_KEY")
 
 # Percorsi ZIP per manuali e problemi
 zip_path_m = "faiss_manual_index.zip"
@@ -30,20 +32,6 @@ embedding_model = HuggingFaceEmbeddings(model_name="sentence-transformers/LaBSE"
 manual_vectorstore = FAISS.load_local(faiss_manual_index, embedding_model, allow_dangerous_deserialization=True)
 problems_vectorstore = FAISS.load_local(faiss_problems_index, embedding_model, allow_dangerous_deserialization=True)
 
-# Caricamento del modello GPT-J da Hugging Face
-model_name = "EleutherAI/gpt-j-6B"
-
-# Forza l'uso della CPU
-model = AutoModelForCausalLM.from_pretrained(
-    model_name, 
-    torch_dtype=torch.float32,  # float32 per la CPU
-    device_map={"": "cpu"}  # Specifica CPU come dispositivo
-)
-
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-tokenizer = AutoTokenizer.from_pretrained(model_name)
-model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
-
 # Funzione per la ricerca e il riassunto
 def search_and_summarize(query):
     # Ricerca nei manuali e problemi
@@ -55,12 +43,17 @@ def search_and_summarize(query):
 
     combined_text = f"Manual Results:\n{manual_output}\n\nProblems Results:\n{problems_output}"
 
-    # Generazione del riassunto con GPT-J
+    # Generazione del riassunto con OpenAI
     input_text = f"Riassumi le seguenti informazioni:\n{combined_text}\n\nRiassunto:"
-    inputs = tokenizer(input_text, return_tensors="pt").to("cuda")
-    output = model.generate(inputs.input_ids, max_length=300, temperature=0.7)
-    summary = tokenizer.decode(output[0], skip_special_tokens=True)
-
+    response = openai.Completion.create(
+        engine="text-davinci-003",  # Puoi scegliere un altro modello se preferisci
+        prompt=input_text,
+        max_tokens=300,
+        temperature=0.7
+    )
+
+    summary = response.choices[0].text.strip()
+    
     return manual_output, problems_output, summary
 
 # Interfaccia Gradio
@@ -70,16 +63,17 @@ iface = gr.Interface(
     outputs=[
         gr.Textbox(label="Manual Results"),
         gr.Textbox(label="Issues Results"),
-        gr.Textbox(label="Summary by GPT-J")
+        gr.Textbox(label="Summary by OpenAI")
     ],
     examples=[
         ["How to change the knife?"],
         ["What are the safety precautions for using the machine?"],
         ["How can I get help with the machine?"]
     ],
-    title="Manual Querying System with GPT-J Summarization",
-    description="Enter a question to get information from the manual and the common issues, summarized by GPT-J."
+    title="Manual Querying System with OpenAI Summarization",
+    description="Enter a question to get information from the manual and the common issues, summarized by OpenAI."
 )
 
 # Avvia l'app Gradio
 iface.launch()
+