Add current Agent

app.py

@@ -90,11 +90,16 @@ from codeinterpreterapi import CodeInterpreterSession
 import html2text
 
 from interpreter.code_interpreter import CodeInterpreter
-from interpreter.code_block import CodeBlock
+# from interpreter.code_block import CodeBlock
 
 import regex
 
+
+
 class CodeBlock:
+    '''
+    CodeBlock Class which is able to run in Code Runner
+    '''
     def __init__(self, code):
         self.code = code
         self.output = ""
@@ -115,7 +120,9 @@ code_2 = """
 
 def Code_Runner(code_raw: str):
     # interpreter = CodeInterpreter(language="python", debug_mode=True)
-    code_raw = RemoveIndent(code_raw)
+    global CurrentAgent
+    if CurrentAgent == "Zero Short React 2":
+        code_raw = RemoveIndent(code_raw)
     if '!pip' in code_raw:
         code_raw=code_raw.replace('!pip', 'pip')
         interpreter = CodeInterpreter(language="shell", debug_mode=True)
@@ -278,9 +285,9 @@ class GPTRemote(LLM):
         if 'Action:' in output and 'Observation:' in output:
             output = output.split('Observation:')[0]
 
-        global Choice
+        global CurrentAgent
         # if Choice == "Structured Zero Short Agent":
-        if 1:
+        if CurrentAgent == 'Structured Zero Short Agent':
             try:
                 # temp = output.split('{')[1].split('}')[0:-2]
                 pattern = r'\{((?:[^{}]|(?R))*)\}'
@@ -567,6 +574,7 @@ ListAgentWithRemoteGPT = ['Zero Short React 2','Zero Short Agent 2',
 def SummarizeDoc():
     global vectordb_p
     global Choice
+    global CurrentAgent
     # pinecone.Index(index_name).delete(delete_all=True, namespace='')
     # collection = vectordb_p.get()
     # split_docs = process_documents([metadata['source'] for metadata in collection['metadatas']])
@@ -576,7 +584,7 @@ def SummarizeDoc():
     print(split_docs[tt-1])
     sum_text=""
     try:
-        if Choice in ListAgentWithRemoteGPT:
+        if CurrentAgent in ListAgentWithRemoteGPT:
             sum_chain = load_summarize_chain(GPTfake, chain_type='refine', verbose=True)
         else:
             sum_chain = load_summarize_chain(llm, chain_type='refine', verbose=True)
@@ -1161,9 +1169,12 @@ agent_OPENAI_MULTI = AgentExecutor.from_agent_and_tools(
 # agent.max_execution_time = int(os.getenv("max_iterations"))
 # agent.handle_parsing_errors = True
 # agent.early_stopping_method = "generate"
+global CurrentAgent
+CurrentAgent = ""
 
 def SetAgent(Choice):
     global agent
+    global CurrentAgent
     if Choice =='Zero Short Agent':
         agent = agent_ZEROSHOT_AGENT
         print("Set to:", Choice)
@@ -1191,7 +1202,9 @@ def SetAgent(Choice):
     elif Choice =='Structured Zero Short Agent':
         agent = agent_STRUCTURED_ZEROSHOT_REACT
         print("Set to:", Choice)
-
+    
+    CurrentAgent = Choice
+    return CurrentAgent
 
         
 
@@ -1870,12 +1883,13 @@ def QAQuery_p(question: str):
     global vectordb_p
     global agent
     global Choice
+    global CurrentAgent
     # vectordb = Chroma(persist_directory='db', embedding_function=embeddings)
     retriever = vectordb_p.as_retriever()
     retriever.search_kwargs['k'] = int(os.environ["search_kwargs_k"])
     # retriever.search_kwargs['fetch_k'] = 100
     # if agent == agent_ZEROSHOT_REACT_2 or agent == agent_ZEROSHOT_AGENT_2:
-    if Choice in ListAgentWithRemoteGPT:
+    if CurrentAgent in ListAgentWithRemoteGPT:
         print("--------------- QA with Remote --------------")
         qa = RetrievalQA.from_chain_type(llm=GPTfake, chain_type="stuff", 
                                         retriever=retriever, return_source_documents = True,