Delete chatAgentPdf

chatAgentPdf/utils/ask_human.py

@@ -1,32 +0,0 @@
-""" 
-Custom Langchain tool to ask human
-"""
-
-import time
-
-import streamlit as st
-from langchain.tools.base import BaseTool
-
-
-class CustomAskHumanTool(BaseTool):
-    """Tool that asks user for input."""
-
-    name = "AskHuman"
-    description = """Use this tool if you don't find a specific answer using KendraRetrievalTool.\
-Ask the human to clarify the question or provide the missing information.\
-The input should be a question for the human."""
-
-    def _run(
-        self,
-        query: str,
-        run_manager=None,
-    ) -> str:
-        if "user_answer" not in st.session_state:
-            answer_container = st.chat_message("assistant", avatar="🦜")
-            answer_container.write(query)
-
-            answer = st.text_input("Enter your answer", key="user_answer")
-            while answer == "":
-                time.sleep(1)
-
-        return st.session_state["user_answer"]

chatAgentPdf/utils/model_params.py

@@ -1,51 +0,0 @@
-"""
-Utilities for modeling
-"""
-
-
-def get_model_params(
-    model_id: str,
-    params: dict,
-) -> dict:
-    """
-    Set up a dictionary with model parameters named appropriately for Bedrock
-
-    Parameters
-    ----------
-    model_id : str
-        Model name
-    params : dict
-        Inference parameters
-
-    Returns
-    -------
-    dict
-        _description_
-    """
-
-    model_params = {}
-
-    # name parameters based on the model id
-    if model_id.startswith("amazon"):
-        model_params = {
-            "maxTokenCount": params["answer_length"],
-            "stopSequences": params["stop_words"],
-            "temperature": params["temperature"],
-            "topP": params["top_p"],
-        }
-    elif model_id.startswith("anthropic"):
-        model_params = {
-            "max_tokens_to_sample": params["answer_length"],
-            "stop_sequences": params["stop_words"],
-            "temperature": params["temperature"],
-            "top_p": params["top_p"],
-        }
-    elif model_id.startswith("ai21"):
-        model_params = {
-            "maxTokens": params["answer_length"],
-            "stopSequences": params["stop_words"],
-            "temperature": params["temperature"],
-            "topP": params["top_p"],
-        }
-
-    return model_params

chatAgentPdf/utils/prompts.py

@@ -1,49 +0,0 @@
-""" 
-Custom Langchain prompt templates
-"""
-
-from langchain.prompts import PromptTemplate
-
-
-def create_qa_prompt() -> PromptTemplate:
-    """
-    Prompt for retrieval QA chain
-    """
-
-    template = """\n\nHuman: Use the following pieces of context to answer the question at the end.
-
-{context}
-
-Question: {question}
-
-\n\nAssistant:
-Answer:"""
-
-    return PromptTemplate(template=template, input_variables=["context", "question"])
-
-
-def create_agent_prompt() -> PromptTemplate:
-    """
-    Prompt for the agent
-    """
-
-    prefix = """\n\nHuman: Answer the following questions as best you can. You have access to the following tools:"""
-
-    format_instructions = """Use the following format:
-
-Question: the input question you must answer
-Thought: you should always think about what to do
-Action: the action to take, should be one of [{tool_names}]
-Action Input: the input to the action
-Observation: the result of the action
-... (this Thought/Action/Action Input/Observation can repeat N times)
-Thought: I now know the final answer
-Final Answer: the final answer to the original input question"""
-
-    suffix = """Begin!
-Question: {input}
-\n\nAssistant:
-Thought: {agent_scratchpad}
-"""
-
-    return prefix, format_instructions, suffix