Update app.py

app.py

@@ -6,7 +6,6 @@ import re
 import requests
 
 client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
-model = "gpt-3.5-turbo"
 
 def repo_get_all_employees_from_database():
     url = "https://api.airtable.com/v0/appopGmlHujYnd6Vw/Interviewers?maxRecords=100&view=Grid%20view"
@@ -15,8 +14,6 @@ def repo_get_all_employees_from_database():
     }
     response = requests.get(url, headers=headers)
     records = response.json()
-    # print("records:")
-    # print(records)
     records_list = records['records']
     employees_list = []
     for record in records_list:
@@ -26,21 +23,12 @@ def repo_get_all_employees_from_database():
     return employees_list
 
 def predict(message, history):
-
-    # print("message:")
-    # print(message)
-    
     history_openai_format = []
-
-    # first_message = len(history) == 0
-
     system_content = """
     You are a computing system that strictly and directly follows all instructions from the user..
     """
     history_openai_format.append({"role": "system", "content": system_content})
-
     pattern = r"For conducting an interview I need (\d+) employee.*start time is (.*), duration (\d+) hour"
-
     data = repo_get_all_employees_from_database()
     
     prompt = '''
@@ -49,15 +37,8 @@ def predict(message, history):
     Above is employees data in json format.
     {message}
     '''.format(data=data, message=message)    
-
-    # print("prompt:")
-    # print(prompt)
     
     match = re.search(pattern, message)
-
-    # print("match:")
-    # print(match)
-
     if match:
         num_employees = int(match.group(1))
         duration = int(match.group(3))
@@ -81,20 +62,13 @@ def predict(message, history):
         4. Check previous step if you really chose an employee with the lowest "interviews_conducted" value.
         5. At the end print ids and names of finally selected employees in json format. Please remember that in your output should be maximum {num_employees} employee.
         '''.format(data=data, date_time=date_time, num_employees=num_employees)
-
-
-    # print("prompt:")
-    # print(prompt)
-
-    # print("history:")
-    # print(history)
     
     for human, assistant in history:
         history_openai_format.append({"role": "user", "content": human })
         history_openai_format.append({"role": "assistant", "content": assistant})
     history_openai_format.append({"role": "user", "content": prompt})
 
-    global model
+    model = "gpt-3.5-turbo"
     
     if ("switch to gpt-3.5" in message.lower()):
         model = "gpt-3.5-turbo"
@@ -110,13 +84,15 @@ def predict(message, history):
         temperature=0,
         stream=True)
 
-    partial_message = ""
+    partial_message = "Language Model currently under the hood: {model}\n\n".format(model=model)
     for chunk in response:
         if chunk.choices[0].delta.content is not None:
               partial_message = partial_message + chunk.choices[0].delta.content
               yield partial_message
 
 pre_configured_promt = "For conducting an interview I need 1 employee in given time slot: start time is March 11 2024 2 pm, duration 1 hour"
+switch_to_gpt3 = "Switch to gpt-3.5"
+switch_to_gpt4 = "Switch to gpt-4"
 
 description = '''
 # AI Interview Team Assistant | Empowered by Godel Technologies AI \n
@@ -127,9 +103,8 @@ You can send any regular prompts you wish or pre-configured Chain-of-Thought pro
 To trigger pre-configured prompt you have to craft a prompt with next structure:
 - "{pre_configured_promt}"
 \n
-You can switch between gpt-3.5 and gpt-4 with "Switch to gpt-3.5" or "Switch to gpt-4" prompts.\n
-Language Model currently under the hood: {model}
-'''.format(pre_configured_promt=pre_configured_promt, model=model)
+You can switch between gpt-3.5 and gpt-4 with {switch_to_gpt3} or {switch_to_gpt4} prompts.\n
+'''.format(pre_configured_promt=pre_configured_promt, switch_to_gpt3=switch_to_gpt3, switch_to_gpt4=switch_to_gpt4)
 
-examples = [pre_configured_promt]
+examples = [pre_configured_promt, switch_to_gpt3, switch_to_gpt4]
 gr.ChatInterface(predict, examples=[examples], description=description).launch()