chat with pdf

.gitignore

@@ -0,0 +1,5 @@
+.venv
+.env
+__pycache__
+models
+*.bin

Makefile.venv

@@ -0,0 +1,272 @@
+#
+# SEAMLESSLY MANAGE PYTHON VIRTUAL ENVIRONMENT WITH A MAKEFILE
+#
+# https://github.com/sio/Makefile.venv             v2023.04.17
+#
+#
+# Insert `include Makefile.venv` at the bottom of your Makefile to enable these
+# rules.
+#
+# When writing your Makefile use '$(VENV)/python' to refer to the Python
+# interpreter within virtual environment and '$(VENV)/executablename' for any
+# other executable in venv.
+#
+# This Makefile provides the following targets:
+#   venv
+#       Use this as a dependency for any target that requires virtual
+#       environment to be created and configured
+#   python, ipython
+#       Use these to launch interactive Python shell within virtual environment
+#   shell, bash, zsh
+#       Launch interactive command line shell. "shell" target launches the
+#       default shell Makefile executes its rules in (usually /bin/sh).
+#       "bash" and "zsh" can be used to refer to the specific desired shell.
+#   show-venv
+#       Show versions of Python and pip, and the path to the virtual environment
+#   clean-venv
+#       Remove virtual environment
+#   $(VENV)/executable_name
+#       Install `executable_name` with pip. Only packages with names matching
+#       the name of the corresponding executable are supported.
+#       Use this as a lightweight mechanism for development dependencies
+#       tracking. E.g. for one-off tools that are not required in every
+#       developer's environment, therefore are not included into
+#       requirements.txt or setup.py.
+#       Note:
+#           Rules using such target or dependency MUST be defined below
+#           `include` directive to make use of correct $(VENV) value.
+#       Example:
+#           codestyle: $(VENV)/pyflakes
+#               $(VENV)/pyflakes .
+#       See `ipython` target below for another example.
+#
+# This Makefile can be configured via following variables:
+#   PY
+#       Command name for system Python interpreter. It is used only initially to
+#       create the virtual environment
+#       Default: python3
+#   REQUIREMENTS_TXT
+#       Space separated list of paths to requirements.txt files.
+#       Paths are resolved relative to current working directory.
+#       Default: requirements.txt
+#
+#       Non-existent files are treated as hard dependencies,
+#       recipes for creating such files must be provided by the main Makefile.
+#       Providing empty value (REQUIREMENTS_TXT=) turns off processing of
+#       requirements.txt even when the file exists.
+#   SETUP_PY, SETUP_CFG, PYPROJECT_TOML, VENV_LOCAL_PACKAGE
+#       Space separated list of paths to files that contain build instructions
+#       for local Python packages. Corresponding packages will be installed
+#       into venv in editable mode along with all their dependencies.
+#       Default: setup.py setup.cfg pyproject.toml (whichever present)
+#
+#       Non-existent and empty values are treated in the same way as for REQUIREMENTS_TXT.
+#   WORKDIR
+#       Parent directory for the virtual environment.
+#       Default: current working directory.
+#   VENVDIR
+#       Python virtual environment directory.
+#       Default: $(WORKDIR)/.venv
+#
+# This Makefile was written for GNU Make and may not work with other make
+# implementations.
+#
+#
+# Copyright (c) 2019-2023 Vitaly Potyarkin
+#
+# Licensed under the Apache License, Version 2.0
+#    <http://www.apache.org/licenses/LICENSE-2.0>
+#
+
+
+#
+# Configuration variables
+#
+
+WORKDIR?=.
+VENVDIR?=$(WORKDIR)/.venv
+REQUIREMENTS_TXT?=$(wildcard requirements.txt)  # Multiple paths are supported (space separated)
+SETUP_PY?=$(wildcard setup.py)                  # Multiple paths are supported (space separated)
+SETUP_CFG?=$(foreach s,$(SETUP_PY),$(wildcard $(patsubst %setup.py,%setup.cfg,$(s))))
+PYPROJECT_TOML?=$(wildcard pyproject.toml)
+VENV_LOCAL_PACKAGE?=$(SETUP_PY) $(SETUP_CFG) $(PYPROJECT_TOML)
+MARKER=.initialized-with-Makefile.venv
+
+
+#
+# Python interpreter detection
+#
+
+_PY_AUTODETECT_MSG=Detected Python interpreter: $(PY). Use PY environment variable to override
+
+ifeq (ok,$(shell test -e /dev/null 2>&1 && echo ok))
+NULL_STDERR=2>/dev/null
+else
+NULL_STDERR=2>NUL
+endif
+
+ifndef PY
+_PY_OPTION:=python3
+ifeq (ok,$(shell $(_PY_OPTION) -c "print('ok')" $(NULL_STDERR)))
+PY=$(_PY_OPTION)
+endif
+endif
+
+ifndef PY
+_PY_OPTION:=$(VENVDIR)/bin/python
+ifeq (ok,$(shell $(_PY_OPTION) -c "print('ok')" $(NULL_STDERR)))
+PY=$(_PY_OPTION)
+$(info $(_PY_AUTODETECT_MSG))
+endif
+endif
+
+ifndef PY
+_PY_OPTION:=$(subst /,\,$(VENVDIR)/Scripts/python)
+ifeq (ok,$(shell $(_PY_OPTION) -c "print('ok')" $(NULL_STDERR)))
+PY=$(_PY_OPTION)
+$(info $(_PY_AUTODETECT_MSG))
+endif
+endif
+
+ifndef PY
+_PY_OPTION:=py -3
+ifeq (ok,$(shell $(_PY_OPTION) -c "print('ok')" $(NULL_STDERR)))
+PY=$(_PY_OPTION)
+$(info $(_PY_AUTODETECT_MSG))
+endif
+endif
+
+ifndef PY
+_PY_OPTION:=python
+ifeq (ok,$(shell $(_PY_OPTION) -c "print('ok')" $(NULL_STDERR)))
+PY=$(_PY_OPTION)
+$(info $(_PY_AUTODETECT_MSG))
+endif
+endif
+
+ifndef PY
+define _PY_AUTODETECT_ERR
+Could not detect Python interpreter automatically.
+Please specify path to interpreter via PY environment variable.
+endef
+$(error $(_PY_AUTODETECT_ERR))
+endif
+
+
+#
+# Internal variable resolution
+#
+
+VENV=$(VENVDIR)/bin
+EXE=
+# Detect windows
+ifeq (win32,$(shell $(PY) -c "import __future__, sys; print(sys.platform)"))
+VENV=$(VENVDIR)/Scripts
+EXE=.exe
+endif
+
+touch=touch $(1)
+ifeq (,$(shell command -v touch $(NULL_STDERR)))
+# https://ss64.com/nt/touch.html
+touch=type nul >> $(subst /,\,$(1)) && copy /y /b $(subst /,\,$(1))+,, $(subst /,\,$(1))
+endif
+
+RM?=rm -f
+ifeq (,$(shell command -v $(firstword $(RM)) $(NULL_STDERR)))
+RMDIR:=rd /s /q
+else
+RMDIR:=$(RM) -r
+endif
+
+
+#
+# Virtual environment
+#
+
+.PHONY: venv
+venv: $(VENV)/$(MARKER)
+
+.PHONY: clean-venv
+clean-venv:
+	-$(RMDIR) "$(VENVDIR)"
+
+.PHONY: show-venv
+show-venv: venv
+	@$(VENV)/python -c "import sys; print('Python ' + sys.version.replace('\n',''))"
+	@$(VENV)/pip --version
+	@echo venv: $(VENVDIR)
+
+.PHONY: debug-venv
+debug-venv:
+	@echo "PATH (Shell)=$$PATH"
+	@$(MAKE) --version
+	$(info PATH (GNU Make)="$(PATH)")
+	$(info SHELL="$(SHELL)")
+	$(info PY="$(PY)")
+	$(info REQUIREMENTS_TXT="$(REQUIREMENTS_TXT)")
+	$(info VENV_LOCAL_PACKAGE="$(VENV_LOCAL_PACKAGE)")
+	$(info VENVDIR="$(VENVDIR)")
+	$(info VENVDEPENDS="$(VENVDEPENDS)")
+	$(info WORKDIR="$(WORKDIR)")
+
+
+#
+# Dependencies
+#
+
+ifneq ($(strip $(REQUIREMENTS_TXT)),)
+VENVDEPENDS+=$(REQUIREMENTS_TXT)
+endif
+
+ifneq ($(strip $(VENV_LOCAL_PACKAGE)),)
+VENVDEPENDS+=$(VENV_LOCAL_PACKAGE)
+endif
+
+$(VENV):
+	$(PY) -m venv $(VENVDIR)
+	$(VENV)/python -m pip install --upgrade pip setuptools wheel
+
+$(VENV)/$(MARKER): $(VENVDEPENDS) | $(VENV)
+ifneq ($(strip $(REQUIREMENTS_TXT)),)
+	$(VENV)/pip install $(foreach path,$(REQUIREMENTS_TXT),-r $(path))
+endif
+ifneq ($(strip $(VENV_LOCAL_PACKAGE)),)
+	$(VENV)/pip install $(foreach path,$(sort $(VENV_LOCAL_PACKAGE)),-e $(dir $(path)))
+endif
+	$(call touch,$(VENV)/$(MARKER))
+
+
+#
+# Interactive shells
+#
+
+.PHONY: python
+python: venv
+	exec $(VENV)/python
+
+.PHONY: ipython
+ipython: $(VENV)/ipython
+	exec $(VENV)/ipython
+
+.PHONY: shell
+shell: venv
+	. $(VENV)/activate && exec $(notdir $(SHELL))
+
+.PHONY: bash zsh
+bash zsh: venv
+	. $(VENV)/activate && exec $@
+
+
+#
+# Commandline tools (wildcard rule, executable name must match package name)
+#
+
+ifneq ($(EXE),)
+$(VENV)/%: $(VENV)/%$(EXE) ;
+.PHONY:    $(VENV)/%
+.PRECIOUS: $(VENV)/%$(EXE)
+endif
+
+$(VENV)/%$(EXE): $(VENV)/$(MARKER)
+	$(VENV)/pip install --upgrade $*
+	$(call touch,$@)

app.py

@@ -0,0 +1,104 @@
+import streamlit as st
+from dotenv import load_dotenv
+from PyPDF2 import PdfReader
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.embeddings import OpenAIEmbeddings, HuggingFaceInstructEmbeddings
+from langchain.vectorstores import FAISS
+from langchain.chat_models import ChatOpenAI
+from langchain.memory import ConversationBufferMemory
+from langchain.chains import ConversationalRetrievalChain
+from htmlTemplates import css, bot_template, user_template
+from langchain.llms import HuggingFaceHub
+
+def get_pdf_text(pdf_docs):
+    text = ""
+    for pdf in pdf_docs:
+        pdf_reader = PdfReader(pdf)
+        for page in pdf_reader.pages:
+            text += page.extract_text()
+    return text
+
+
+def get_text_chunks(text):
+    text_splitter = CharacterTextSplitter(
+        separator="\n",
+        chunk_size=1000,
+        chunk_overlap=200,
+        length_function=len
+    )
+    chunks = text_splitter.split_text(text)
+    return chunks
+
+
+def get_vectorstore(text_chunks):
+    embeddings = OpenAIEmbeddings()
+    # embeddings = HuggingFaceInstructEmbeddings(model_name="hkunlp/instructor-xl")
+    vectorstore = FAISS.from_texts(texts=text_chunks, embedding=embeddings)
+    return vectorstore
+
+
+def get_conversation_chain(vectorstore):
+    llm = ChatOpenAI()
+    # llm = HuggingFaceHub(repo_id="google/flan-t5-xxl", model_kwargs={"temperature":0.5, "max_length":512})
+
+    memory = ConversationBufferMemory(
+        memory_key='chat_history', return_messages=True)
+    conversation_chain = ConversationalRetrievalChain.from_llm(
+        llm=llm,
+        retriever=vectorstore.as_retriever(),
+        memory=memory
+    )
+    return conversation_chain
+
+
+def handle_userinput(user_question):
+    response = st.session_state.conversation({'question': user_question})
+    st.session_state.chat_history = response['chat_history']
+
+    for i, message in enumerate(st.session_state.chat_history):
+        if i % 2 == 0:
+            st.write(user_template.replace(
+                "{{MSG}}", message.content), unsafe_allow_html=True)
+        else:
+            st.write(bot_template.replace(
+                "{{MSG}}", message.content), unsafe_allow_html=True)
+
+
+def main():
+    load_dotenv()
+    st.set_page_config(page_title="Chat with multiple PDFs",
+                       page_icon=":books:")
+    st.write(css, unsafe_allow_html=True)
+
+    if "conversation" not in st.session_state:
+        st.session_state.conversation = None
+    if "chat_history" not in st.session_state:
+        st.session_state.chat_history = None
+
+    st.header("Chat with multiple PDFs :books:")
+    user_question = st.text_input("Ask a question about your documents:")
+    if user_question:
+        handle_userinput(user_question)
+
+    with st.sidebar:
+        st.subheader("Your documents")
+        pdf_docs = st.file_uploader(
+            "Upload your PDFs here and click on 'Process'", accept_multiple_files=True)
+        if st.button("Process"):
+            with st.spinner("Processing"):
+                # get pdf text
+                raw_text = get_pdf_text(pdf_docs)
+
+                # get the text chunks
+                text_chunks = get_text_chunks(raw_text)
+
+                # create vector store
+                vectorstore = get_vectorstore(text_chunks)
+
+                # create conversation chain
+                st.session_state.conversation = get_conversation_chain(
+                    vectorstore)
+
+
+if __name__ == '__main__':
+    main()

htmlTemplates.py

@@ -0,0 +1,44 @@
+css = '''
+<style>
+.chat-message {
+    padding: 1.5rem; border-radius: 0.5rem; margin-bottom: 1rem; display: flex
+}
+.chat-message.user {
+    background-color: #2b313e
+}
+.chat-message.bot {
+    background-color: #475063
+}
+.chat-message .avatar {
+  width: 20%;
+}
+.chat-message .avatar img {
+  max-width: 78px;
+  max-height: 78px;
+  border-radius: 50%;
+  object-fit: cover;
+}
+.chat-message .message {
+  width: 80%;
+  padding: 0 1.5rem;
+  color: #fff;
+}
+'''
+
+bot_template = '''
+<div class="chat-message bot">
+    <div class="avatar">
+        <img src="https://i.ibb.co/cN0nmSj/Screenshot-2023-05-28-at-02-37-21.png" style="max-height: 78px; max-width: 78px; border-radius: 50%; object-fit: cover;">
+    </div>
+    <div class="message">{{MSG}}</div>
+</div>
+'''
+
+user_template = '''
+<div class="chat-message user">
+    <div class="avatar">
+        <img src="https://i.ibb.co/rdZC7LZ/Photo-logo-1.png">
+    </div>    
+    <div class="message">{{MSG}}</div>
+</div>
+'''

makefile

@@ -0,0 +1,14 @@
+
+
+PYTHON_BINARY = $(VENV)/python
+STREAMLIT_BINARY = $(VENV)/streamlit
+
+
+re-venv: clean-venv venv
+
+run: venv
+	$(PYTHON_BINARY) app.py
+	$(STREAMLIT_BINARY) run app.py
+
+
+include Makefile.venv

requirements.txt

@@ -0,0 +1,21 @@
+langchain
+openai
+Streamlit
+huggingface_hub
+python-dotenv
+watchdog
+tiktoken
+pinecone-client
+joblib
+pandas
+scikit-learn
+sentence_transformers
+uvicorn
+ctransformers
+fastapi
+ipykernel
+python-box
+transformers
+PyPDF2
+faiss-cpu
+altair