Create 💊drug_and_medicine_chatbot.py

pages/💊drug_and_medicine_chatbot.py

@@ -0,0 +1,331 @@
+#Import the necessary libraries
+import utils
+import streamlit as st
+from streaming import StreamHandler
+
+from langchain_openai import ChatOpenAI
+from langchain.chains.conversation.base import ConversationChain
+from langchain.memory.buffer import ConversationBufferMemory
+from langchain.memory import ConversationSummaryMemory
+from langchain.chains.conversational_retrieval.base import ConversationalRetrievalChain
+from langchain.prompts import PromptTemplate
+from langchain_community.document_loaders import TextLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_chroma import Chroma
+from langchain_openai import OpenAIEmbeddings
+
+st.set_page_config(page_title="Sickle cell chatbot", page_icon="🧠")
+st.header('Sickle cell chatbot')
+st.write('Enhancing Chatbot Interactions through sickle cell training')
+
+class ContextChatbot:
+
+    def __init__(self):
+        self.openai_model = utils.configure_openai()
+
+    @st.cache_resource
+    def setup_chain(_self):
+        file_paths = [
+            './Docs/Prenatal Diagonosis.txt',
+            './Docs/Sickle Cell Anaemia Yoruba.txt',
+            './Docs/Sickle Cell Anemia.txt',
+            './Docs/Sickle cell in Africa.txt',
+            './Docs/sickle-cell-disease-report.txt',
+            './Docs/Blood Basics.txt',
+            './Docs/Sickle Cell Basic Info.txt',
+            './Docs/Manging Sicklecell.txt',
+            './Docs/SickleCell_nutrition.txt',
+            './Docs/Lifestyle Advice.txt'
+        ]
+
+        # Load documents
+        documents = []
+        for path in file_paths:
+            try:
+                loader = TextLoader(path)
+                documents.extend(loader.load())
+                print(f"Successfully loaded: {path}")
+            except Exception as e:
+                print(f"Error loading {path}: {e}")
+
+                # Check if any documents were loaded
+        if not documents:
+            raise RuntimeError("No documents were loaded. Please check the file paths and file contents.")
+        text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=650, chunk_overlap=60, add_start_index=True
+        )
+        openai_embedding_model = OpenAIEmbeddings(
+            model="text-embedding-ada-002",
+        )
+        all_splits = text_splitter.split_documents(documents)
+        vectorstore = Chroma.from_documents(documents=all_splits, embedding=openai_embedding_model, persist_directory='./Vectorstore/')
+        prompt_template = """
+        Use the following pieces of context to answer the user's question about sickle cell.
+        If you don't know the answer, just say that you don't know, don't try to make up an answer.
+        ----------------
+        {context}
+        Question: {question}
+        """
+
+        PROMPT = PromptTemplate(
+            template=prompt_template, input_variables=["context", "question"]
+        )
+        chain_type_kwargs = {"prompt": PROMPT}
+        summarizer = ChatOpenAI(model_name= "gpt-4o", temperature=0, streaming=True)
+        llm = ChatOpenAI(model_name=_self.openai_model, temperature=0, streaming=True)
+
+        chat_history = []
+
+        qa = ConversationalRetrievalChain.from_llm(
+            llm = llm,
+            chain_type = "stuff",
+            memory = ConversationSummaryMemory(llm = summarizer, memory_key='chat_history', input_key='question', output_key= 'answer', return_messages=True),
+            retriever = vectorstore.as_retriever(k = 5, search_type="mmr"),
+            return_source_documents=True,
+            combine_docs_chain_kwargs=chain_type_kwargs
+        )
+        return qa
+
+    @utils.enable_chat_history
+    def main(self):
+        chain = self.setup_chain()
+        user_query = st.chat_input(placeholder="Ask me anything!")
+        if user_query:
+            utils.display_msg(user_query, 'user')
+            with st.chat_message("assistant"):
+                st_cb = StreamHandler(st.empty())
+                result = chain.invoke(
+                    {"question":user_query},
+                    {"callbacks": [st_cb]}
+                )
+                #response = ""
+                #for source in result['source_documents']:
+                    #response += source.metadata['source']
+                    #response += "Content: " + source.page_content
+                response = result["answer"]
+                st.session_state.messages.append({"role": "assistant", "content": response})
+
+if __name__ == "__main__":
+    obj = ContextChatbot()
+    obj.main()
+#SickleCell#Import the necessary libraries
+import utils
+import streamlit as st
+from streaming import StreamHandler
+
+from langchain_openai import ChatOpenAI
+from langchain.chains.conversation.base import ConversationChain
+from langchain.memory.buffer import ConversationBufferMemory
+from langchain.memory import ConversationSummaryMemory
+from langchain.chains.conversational_retrieval.base import ConversationalRetrievalChain
+from langchain.prompts import PromptTemplate
+from langchain_community.document_loaders import TextLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_chroma import Chroma
+from langchain_openai import OpenAIEmbeddings
+
+st.set_page_config(page_title="Drug and Medicine chatbot", page_icon="💊")
+st.header('Sickle cell chatbot')
+st.write('Enhancing Chatbot Interactions through sickle cell training')
+
+class ContextChatbot:
+
+    def __init__(self):
+        self.openai_model = utils.configure_openai()
+
+    @st.cache_resource
+    def setup_chain(_self):
+        file_paths = [
+            './Docs/Prenatal Diagonosis.txt',
+            './Docs/Sickle Cell Anaemia Yoruba.txt',
+            './Docs/Sickle Cell Anemia.txt',
+            './Docs/Sickle cell in Africa.txt',
+            './Docs/sickle-cell-disease-report.txt',
+            './Docs/Blood Basics.txt',
+            './Docs/Sickle Cell Basic Info.txt',
+            './Docs/Manging Sicklecell.txt',
+            './Docs/SickleCell_nutrition.txt',
+            './Docs/Lifestyle Advice.txt'
+        ]
+
+        # Load documents
+        documents = []
+        for path in file_paths:
+            try:
+                loader = TextLoader(path)
+                documents.extend(loader.load())
+                print(f"Successfully loaded: {path}")
+            except Exception as e:
+                print(f"Error loading {path}: {e}")
+
+                # Check if any documents were loaded
+        if not documents:
+            raise RuntimeError("No documents were loaded. Please check the file paths and file contents.")
+        text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=650, chunk_overlap=60, add_start_index=True
+        )
+        openai_embedding_model = OpenAIEmbeddings(
+            model="text-embedding-ada-002",
+        )
+        all_splits = text_splitter.split_documents(documents)
+        vectorstore = Chroma.from_documents(documents=all_splits, embedding=openai_embedding_model, persist_directory='./Vectorstore/')
+        prompt_template = """
+        Use the following pieces of context to answer the user's question about sickle cell.
+        If you don't know the answer, just say that you don't know, don't try to make up an answer.
+        ----------------
+        {context}
+        Question: {question}
+        """
+
+        PROMPT = PromptTemplate(
+            template=prompt_template, input_variables=["context", "question"]
+        )
+        chain_type_kwargs = {"prompt": PROMPT}
+        summarizer = ChatOpenAI(model_name= "gpt-4o", temperature=0, streaming=True)
+        llm = ChatOpenAI(model_name=_self.openai_model, temperature=0, streaming=True)
+
+        chat_history = []
+
+        qa = ConversationalRetrievalChain.from_llm(
+            llm = llm,
+            chain_type = "stuff",
+            memory = ConversationSummaryMemory(llm = summarizer, memory_key='chat_history', input_key='question', output_key= 'answer', return_messages=True),
+            retriever = vectorstore.as_retriever(k = 5, search_type="mmr"),
+            return_source_documents=True,
+            combine_docs_chain_kwargs=chain_type_kwargs
+        )
+        return qa
+
+    @utils.enable_chat_history
+    def main(self):
+        chain = self.setup_chain()
+        user_query = st.chat_input(placeholder="Ask me anything!")
+        if user_query:
+            utils.display_msg(user_query, 'user')
+            with st.chat_message("assistant"):
+                st_cb = StreamHandler(st.empty())
+                result = chain.invoke(
+                    {"question":user_query},
+                    {"callbacks": [st_cb]}
+                )
+                #response = ""
+                #for source in result['source_documents']:
+                    #response += source.metadata['source']
+                    #response += "Content: " + source.page_content
+                response = result["answer"]
+                st.session_state.messages.append({"role": "assistant", "content": response})
+
+if __name__ == "__main__":
+    obj = ContextChatbot()
+    obj.main()
+#SickleCell#Import the necessary libraries
+import utils
+import streamlit as st
+from streaming import StreamHandler
+
+from langchain_openai import ChatOpenAI
+from langchain.chains.conversation.base import ConversationChain
+from langchain.memory.buffer import ConversationBufferMemory
+from langchain.memory import ConversationSummaryMemory
+from langchain.chains.conversational_retrieval.base import ConversationalRetrievalChain
+from langchain.prompts import PromptTemplate
+from langchain_community.document_loaders import TextLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain_chroma import Chroma
+from langchain_openai import OpenAIEmbeddings
+
+st.set_page_config(page_title="Sickle cell chatbot", page_icon="🧠")
+st.header('Sickle cell chatbot')
+st.write('Enhancing Chatbot Interactions through sickle cell training')
+
+class ContextChatbot:
+
+    def __init__(self):
+        self.openai_model = utils.configure_openai()
+
+    @st.cache_resource
+    def setup_chain(_self):
+        file_paths = [
+            './Docs/Prenatal Diagonosis.txt',
+            './Docs/Sickle Cell Anaemia Yoruba.txt',
+            './Docs/Sickle Cell Anemia.txt',
+            './Docs/Sickle cell in Africa.txt',
+            './Docs/sickle-cell-disease-report.txt',
+            './Docs/Blood Basics.txt',
+            './Docs/Sickle Cell Basic Info.txt',
+            './Docs/Manging Sicklecell.txt',
+            './Docs/SickleCell_nutrition.txt',
+            './Docs/Lifestyle Advice.txt'
+        ]
+
+        # Load documents
+        documents = []
+        for path in file_paths:
+            try:
+                loader = TextLoader(path)
+                documents.extend(loader.load())
+                print(f"Successfully loaded: {path}")
+            except Exception as e:
+                print(f"Error loading {path}: {e}")
+
+                # Check if any documents were loaded
+        if not documents:
+            raise RuntimeError("No documents were loaded. Please check the file paths and file contents.")
+        text_splitter = RecursiveCharacterTextSplitter(
+            chunk_size=650, chunk_overlap=60, add_start_index=True
+        )
+        openai_embedding_model = OpenAIEmbeddings(
+            model="text-embedding-ada-002",
+        )
+        all_splits = text_splitter.split_documents(documents)
+        vectorstore = Chroma.from_documents(documents=all_splits, embedding=openai_embedding_model, persist_directory='./Vectorstore/')
+        prompt_template = """
+        Use the following pieces of context to answer the user's question about sickle cell.
+        If you don't know the answer, just say that you don't know, don't try to make up an answer.
+        ----------------
+        {context}
+        Question: {question}
+        """
+
+        PROMPT = PromptTemplate(
+            template=prompt_template, input_variables=["context", "question"]
+        )
+        chain_type_kwargs = {"prompt": PROMPT}
+        summarizer = ChatOpenAI(model_name= "gpt-4o", temperature=0, streaming=True)
+        llm = ChatOpenAI(model_name=_self.openai_model, temperature=0, streaming=True)
+
+        chat_history = []
+
+        qa = ConversationalRetrievalChain.from_llm(
+            llm = llm,
+            chain_type = "stuff",
+            memory = ConversationSummaryMemory(llm = summarizer, memory_key='chat_history', input_key='question', output_key= 'answer', return_messages=True),
+            retriever = vectorstore.as_retriever(k = 5, search_type="mmr"),
+            return_source_documents=True,
+            combine_docs_chain_kwargs=chain_type_kwargs
+        )
+        return qa
+
+    @utils.enable_chat_history
+    def main(self):
+        chain = self.setup_chain()
+        user_query = st.chat_input(placeholder="Ask me anything!")
+        if user_query:
+            utils.display_msg(user_query, 'user')
+            with st.chat_message("assistant"):
+                st_cb = StreamHandler(st.empty())
+                result = chain.invoke(
+                    {"question":user_query},
+                    {"callbacks": [st_cb]}
+                )
+                #response = ""
+                #for source in result['source_documents']:
+                    #response += source.metadata['source']
+                    #response += "Content: " + source.page_content
+                response = result["answer"]
+                st.session_state.messages.append({"role": "assistant", "content": response})
+
+if __name__ == "__main__":
+    obj = ContextChatbot()
+    obj.main()
+#SickleCell