Update app.py

app.py

@@ -15,8 +15,31 @@ tokenizer = AutoTokenizer.from_pretrained(model_name)
 # Asignar el eos_token como pad_token
 tokenizer.pad_token = tokenizer.eos_token
 
+# Upload CSV file
+uploaded_file = st.file_uploader("Upload a CSV file", type=["csv"])
+#print("Query: ", query)
+
+df = pd.DataFrame(uploaded_file)
+job_titles = df['job_title'].tolist()
+query="aspiring human resources specialist"
+
 # Texto de entrada para la generación
-input_text = "Tu texto aquí"
+input_text = (
+    f"You are an AI assistant. You have a list of job titles and a search query.\n"
+    f"Your task is to rank these job titles by their semantic similarity to the given query. "
+    f"Please provide the ranking from most relevant to least relevant. "
+    f"Do not calculate cosine similarity; instead, focus on understanding the semantic relevance of each job title to the query.\n"
+    f"\n"
+    f"Format your response like this:\n"
+    f"1. [Most Relevant Job Title]\n"
+    f"2. [Second Most Relevant Job Title]\n"
+    f"...\n"
+    f"N. [Least Relevant Job Title]\n"
+    f"\n"
+    f"Query: \"{query}\"\n"
+    f"Job Titles: {job_titles}\n"
+)
+
 inputs = tokenizer(input_text, return_tensors="pt", padding=True)
 
 # Establecer el id del token de padding