Upload folder using huggingface_hub

README.md

@@ -1,12 +1,6 @@
 ---
 title: Gradio
-emoji: 🐢
-colorFrom: yellow
-colorTo: indigo
+app_file: gradio_llm_example.py
 sdk: gradio
-sdk_version: 3.46.1
-app_file: app.py
-pinned: false
+sdk_version: 3.46.0
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

flagged/log.csv

@@ -0,0 +1,2 @@
+input_file,Error,flag,username,timestamp
+,Error occurred while writing the file: 'Textbox' object has no attribute 'configure',,,2023-10-04 11:29:37.852423

gradio_app.py

@@ -0,0 +1,205 @@
+import gradio as gr
+import numpy as np
+import random
+# import torch
+# from langchain import HuggingFacePipeline
+# from langchain.chains import LLMChain, RetrievalQA
+# from langchain.document_loaders import (
+#     DirectoryLoader,
+#     PyPDFLoader,
+#     TextLoader,
+#     UnstructuredPDFLoader,
+# )
+# from langchain.embeddings import HuggingFaceEmbeddings, LlamaCppEmbeddings
+# from langchain.llms import LlamaCpp
+# from langchain.prompts import PromptTemplate
+# from langchain.text_splitter import (
+#     CharacterTextSplitter,
+#     RecursiveCharacterTextSplitter,
+# )
+# from langchain.vectorstores import Chroma
+# from PIL import Image
+
+
+def file_upload(input_file):
+    # Process the uploaded file
+    if input_file is not None:
+        # Save the uploaded file or perform any desired operations
+        file_path = "/tmp/file.pdf"
+        content = input_file.read()
+        try:
+            with open(file_path, 'wb') as file:
+                file.write(content)
+            return {error_box: gr.Textbox(label="Completed",
+                                          value=f"File uploaded successfully in {file_path}.", visible=True)}
+        except Exception as e:
+            return {error_box: gr.Textbox(label="Error",
+                                          value=f"Error occurred while writing the file: {e}", visible=True)}
+
+
+def respond(message, chat_history):
+        #No LLM here, just respond with a random pre-made message
+        bot_message = random.choice(["Tell me more about it", 
+                                     "Cool, but I'm not interested", 
+                                     "Hmmmm, ok then"]) 
+        chat_history.append((message, bot_message))
+        return "", chat_history
+
+
+# Gradio interface
+def qa_bot(pdf_file, question):
+    texts = load_docs(pdf_file)
+    model = setup_dbqa(texts)
+    answer = model({'query': question})
+    return f"Question: {answer['query']}\nAnswer: {answer['result']}\nSource documents: {answer['source_documents']}"
+
+
+# Helper function to load documents from PDF files
+def load_docs(file_path):
+    loader = DirectoryLoader(file_path,
+                             glob="*.pdf",
+                             loader_cls=UnstructuredPDFLoader)
+    documents = loader.load()
+
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000,
+                                                   chunk_overlap=0,
+                                                   length_function=len,)
+    texts = text_splitter.split_documents(documents)
+    return texts
+
+
+# Helper function to set up the question-answering model
+def setup_dbqa(texts):
+    print("Setting up DBQA ...")
+    llm = HuggingFacePipeline.from_model_id(
+        model_id="NousResearch/Llama-2-13b-chat-hf",
+        task="text-generation",
+        model_kwargs={
+            "max_length": 1500, "load_in_8bit": True},
+    )
+
+    embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L6-v2',
+                                       model_kwargs={'device': 'cpu'})
+
+    vectorstore = Chroma.from_documents(texts, embeddings, persist_directory="vectorstore")
+
+    prompt = set_qa_prompt()
+
+    return build_retrieval_qa(llm, prompt, vectorstore)
+
+
+
+def set_qa_prompt():
+    # set prompt template
+    prompt_template = """<s>[INST] <<SYS>> Use the following pieces of context closed between $ to answer the question closed between |. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
+    ${context}$ <</SYS>>
+    Question: |{question}|
+    Answer:[/INST]</s>"""
+    prompt = PromptTemplate(
+        template=prompt_template, input_variables=["context", "question"]
+    )
+    return prompt
+
+
+# Build RetrievalQA object
+
+def build_retrieval_qa(_llm, _prompt, _vectorstore):
+    dbqa = RetrievalQA.from_chain_type(llm=_llm,
+                                       chain_type='stuff',
+                                       retriever=_vectorstore.as_retriever(search_kwargs={'k': 3}),
+                                       return_source_documents=True,
+                                       chain_type_kwargs={'prompt': _prompt})
+    return dbqa
+
+
+
+
+
+if __name__ == "__main__":
+
+    # How to RUN code ==> gradio gradio_app.py 
+    gr.themes.builder()
+
+    # # Define text and title information 
+    # title1 = "## QA App"
+
+    # title2 = " ## Gradio QA Bot"
+
+    # intro = """
+    #     Welcome! This is not just any bot, it's a special one equipped with state-of-the-art natural language processing capabilities, and ready to answer your queries.
+
+    #     Ready to explore? Let's get started!
+
+    #     * Step 1: Upload a PDF document.
+    #     * Step 2: Type in a question related to your document's content.
+    #     * Step 3: Get your answer!
+
+    #     Push clear cache before uploading a new doc!
+    #     """
+
+    # about = """ 
+    #     ## About
+    #     This app is an LLM-powered chatbot built using:
+    #     - [Streamlit](<https://streamlit.io/>)
+    #     - [HugChat](<https://github.com/Soulter/hugging-chat-api>)
+    #     - Chat Model = llama2-chat-hf 7B 
+    #     - Retreiver model = all-MiniLM-L6-v2
+
+    #     💡 Note: No API key required!
+    #     """
+
+
+    # # Define theme ==> see gr.themes.builder()
+    # theme = gr.themes.Soft(
+    #     primary_hue="green",
+    #     secondary_hue="blue",
+    #     neutral_hue="indigo"
+    #     ).set(
+    #     background_fill_primary='*primary_50',
+    #     shadow_drop='*shadow_spread',
+    #     button_border_width='*block_border_width',
+    #     button_border_width_dark='*block_label_border_width'
+    # )
+
+
+    # with gr.Blocks(theme=theme) as demo:
+    #     with gr.Row():
+    #         with gr.Column(scale=2, min_width=400):
+    #             title1_gr= gr.Markdown(title1)
+    #             intro_gr = gr.Markdown(intro)
+    #             # Create a Gradio interface with a file upload input
+    #             error_box = gr.Textbox(label="Error", visible=False)
+    #             # upload_button = gr.Interface(fn=file_upload,
+    #             #                 inputs=gr.File(),
+    #             #                 outputs=error_box,
+    #             #                 description="Drag and drop your document here")
+    #             upload_button = gr.UploadButton("Drag and drop your document here", 
+    #                                             size="lg", scale=3, min_width=240,
+    #                                             file_types=["pdf"])
+    #             upload_button.upload(file_upload, upload_button, error_box)
+
+    #         with gr.Column(scale=2, min_width=800):
+    #             title2_gr = gr.Markdown(title2)
+
+    #             chatbot = gr.Chatbot(label="Bot", height=500) 
+    #             msg = gr.Textbox(label="User", placeholder="Ask a question about the uploaded PDF document.")
+    #             chatbot_btn = gr.Button("Submit")
+    #             clear = gr.ClearButton(components=[msg, chatbot], value="Clear console")
+    #             chatbot_btn.click(respond, inputs=[msg, chatbot], outputs=[msg, chatbot])
+
+
+    #         with gr.Column(scale=3, min_width=600):
+    #             with gr.Row():
+    #                 about_gr = gr.Markdown(about)  
+    #                 logo_gr = gr.Markdown(""" </br> </br>
+    #                                     <img src="file/logo_neovision.png" alt="logo" style="width:600px;"/>""")  
+    #                 # gr.Image("./logo_neovision.png")
+
+
+    # gr.close_all()
+    # demo.launch(share=True, enable_queue=True)
+
+
+
+
+

gradio_cached_examples/16/component 0/tmp95ia8keq.json

@@ -0,0 +1 @@
+[["Am I cool?", "Yes"]]

gradio_cached_examples/16/component 0/tmpf1gyji4c.json

@@ -0,0 +1 @@
+[["Are tomatoes vegetables?", "Yes"]]

gradio_cached_examples/16/component 0/tmpleed6aum.json

@@ -0,0 +1 @@
+[["Hello", "Ask me anything!"]]

gradio_cached_examples/16/log.csv

@@ -0,0 +1,4 @@
+component 0,flag,username,timestamp
+/home/tamara/Documents/Gradio/gradio_cached_examples/16/component 0/tmpleed6aum.json,,,2023-10-03 13:53:06.546567
+/home/tamara/Documents/Gradio/gradio_cached_examples/16/component 0/tmp95ia8keq.json,,,2023-10-03 13:53:06.547644
+/home/tamara/Documents/Gradio/gradio_cached_examples/16/component 0/tmpf1gyji4c.json,,,2023-10-03 13:53:06.548535

gradio_cached_examples/35/component 0/tmphlap_ssj.json

@@ -0,0 +1 @@
+[["Are tomatoes vegetables?", "Yes"]]

gradio_cached_examples/35/component 0/tmpqmrmr545.json

@@ -0,0 +1 @@
+[["Hello", "Ask me anything!"]]

gradio_cached_examples/35/component 0/tmpu4y3h8za.json

@@ -0,0 +1 @@
+[["Am I cool?", "Yes"]]

gradio_cached_examples/35/log.csv

@@ -0,0 +1,4 @@
+component 0,flag,username,timestamp
+/home/tamara/Documents/Gradio/gradio_cached_examples/35/component 0/tmpqmrmr545.json,,,2023-10-03 13:52:57.596267
+/home/tamara/Documents/Gradio/gradio_cached_examples/35/component 0/tmpu4y3h8za.json,,,2023-10-03 13:52:57.596899
+/home/tamara/Documents/Gradio/gradio_cached_examples/35/component 0/tmphlap_ssj.json,,,2023-10-03 13:52:57.597399

gradio_cached_examples/41/component 0/tmp1gnhmzn0.json

@@ -0,0 +1 @@
+[["Are tomatoes vegetables?", "Yes"]]

gradio_cached_examples/41/component 0/tmp9r3rafm7.json

@@ -0,0 +1 @@
+[["Am I cool?", "Yes"]]

gradio_cached_examples/41/component 0/tmphbjuw0z9.json

@@ -0,0 +1 @@
+[["Hello", "Ask me anything!"]]

gradio_cached_examples/41/log.csv

@@ -0,0 +1,4 @@
+component 0,flag,username,timestamp
+/home/tamara/Documents/Gradio/gradio_cached_examples/41/component 0/tmphbjuw0z9.json,,,2023-10-03 14:12:40.387069
+/home/tamara/Documents/Gradio/gradio_cached_examples/41/component 0/tmp9r3rafm7.json,,,2023-10-03 14:12:40.388121
+/home/tamara/Documents/Gradio/gradio_cached_examples/41/component 0/tmp1gnhmzn0.json,,,2023-10-03 14:12:40.388983

gradio_cached_examples/54/component 0/tmpaz_c2ond.json

@@ -0,0 +1 @@
+[["Hello", "Ask me anything!"]]

gradio_cached_examples/54/component 0/tmpnl_8qi5t.json

@@ -0,0 +1 @@
+[["Am I cool?", "Yes"]]

gradio_cached_examples/54/component 0/tmpo6iaiydn.json

@@ -0,0 +1 @@
+[["Are tomatoes vegetables?", "Yes"]]

gradio_cached_examples/54/log.csv

@@ -0,0 +1,4 @@
+component 0,flag,username,timestamp
+/home/tamara/Documents/Gradio/gradio_cached_examples/54/component 0/tmpaz_c2ond.json,,,2023-10-03 13:52:57.635762
+/home/tamara/Documents/Gradio/gradio_cached_examples/54/component 0/tmpnl_8qi5t.json,,,2023-10-03 13:52:57.636386
+/home/tamara/Documents/Gradio/gradio_cached_examples/54/component 0/tmpo6iaiydn.json,,,2023-10-03 13:52:57.636821

gradio_cached_examples/60/component 0/tmp8z177n1e.json

@@ -0,0 +1 @@
+[["Am I cool?", "Yes"]]

gradio_cached_examples/60/component 0/tmpcjfuu9nz.json

@@ -0,0 +1 @@
+[["Are tomatoes vegetables?", "Yes"]]

gradio_cached_examples/60/component 0/tmpzem5dzus.json

@@ -0,0 +1 @@
+[["Hello", "Ask me anything!"]]

gradio_cached_examples/60/log.csv

@@ -0,0 +1,4 @@
+component 0,flag,username,timestamp
+/home/tamara/Documents/Gradio/gradio_cached_examples/60/component 0/tmpzem5dzus.json,,,2023-10-03 14:12:40.417087
+/home/tamara/Documents/Gradio/gradio_cached_examples/60/component 0/tmp8z177n1e.json,,,2023-10-03 14:12:40.418079
+/home/tamara/Documents/Gradio/gradio_cached_examples/60/component 0/tmpcjfuu9nz.json,,,2023-10-03 14:12:40.418965

gradio_cached_examples/79/component 0/tmp00bsbvnd.json

@@ -0,0 +1 @@
+[["Hello", "Ask me anything!"]]

gradio_cached_examples/79/component 0/tmp3b5s5lev.json

@@ -0,0 +1 @@
+[["Am I cool?", "Yes"]]

gradio_cached_examples/79/component 0/tmpmasy_kj7.json

@@ -0,0 +1 @@
+[["Are tomatoes vegetables?", "Yes"]]

gradio_cached_examples/79/log.csv

@@ -0,0 +1,4 @@
+component 0,flag,username,timestamp
+/home/tamara/Documents/Gradio/gradio_cached_examples/79/component 0/tmp00bsbvnd.json,,,2023-10-03 14:12:43.671670
+/home/tamara/Documents/Gradio/gradio_cached_examples/79/component 0/tmp3b5s5lev.json,,,2023-10-03 14:12:43.672661
+/home/tamara/Documents/Gradio/gradio_cached_examples/79/component 0/tmpmasy_kj7.json,,,2023-10-03 14:12:43.673551

gradio_llm_example.py

@@ -0,0 +1,162 @@
+import gradio as gr
+import numpy as np
+import random
+import time
+import os
+import shutil
+import codecs
+# How to RUN code ==> gradio gradio_llm_example.py 
+
+
+# Define text and title information 
+title1 = "##  </br> </br> </br> 🤗💬 QA App"
+
+title2 = " ##  </br> </br> </br> Gradio QA Bot"
+
+intro = """ Welcome! This is not just any bot, it's a special one equipped with state-of-the-art natural language processing capabilities, and ready to answer your queries.
+
+
+    Ready to explore? Let's get started!
+  
+    
+    * Step 1: Upload a PDF document.
+    * Step 2: Type in a question related to your document's content.
+    * Step 3: Get your answer!
+ 
+    
+    Push clear cache before uploading a new doc!
+
+    
+    """
+
+about = """ 
+    ## </br> About
+    This app is an LLM-powered chatbot built using:
+    - [Streamlit](<https://streamlit.io/>)
+    - [HugChat](<https://github.com/Soulter/hugging-chat-api>)
+    - Chat Model = llama2-chat-hf 7B 
+    - Retreiver model = all-MiniLM-L6-v2
+
+    </br> 
+    💡 Note: No API key required!
+    
+     </br> 
+    Made with ❤️ by us
+     """
+
+
+# Define theme ==> see gr.themes.builder()
+theme = gr.themes.Soft(
+    primary_hue="emerald",
+    secondary_hue="emerald",
+    neutral_hue="slate",
+).set(
+    body_background_fill_dark='*primary_50',
+    shadow_drop='*shadow_spread',
+    button_border_width='*block_border_width',
+    button_border_width_dark='*block_label_border_width'
+)
+
+
+def upload_file(files_obj):
+    """ Upload several files from drag and drop, and save them in local temp folder
+    files_obj (type:list) : list of tempfile._TemporaryFileWrapper 
+    return checkbox to display uploaded documents """
+    # Create local copy
+    temp_file_path = "./temp"
+    if not os.path.exists(temp_file_path):
+        os.makedirs(temp_file_path)
+    # Save each file among list of given files
+    file_name_list = list()
+    for file_obj in files_obj :
+        file_name = os.path.basename(file_obj.name)
+        file_name_list.append(file_name)
+        shutil.copyfile(file_obj.name, os.path.join(temp_file_path, file_name))
+    return {uploaded_check : gr.Radio(choices=file_name_list, visible=True),
+            choose_btn : gr.Button(value="Choose", visible=True)}
+
+
+def read_content(file_name):
+    print(file_name, type(file_name))
+    temp_file_path = "./temp"
+    file_path = os.path.join(temp_file_path, file_name)
+    with open(file_path, "rb") as file:
+        try:
+            content = file.read()
+            print(content)
+            print(codecs.decode(content, 'utf-8'))
+            return {error_box: gr.Textbox(value=f"File ready to be used. \n You can ask a question about the uploaded PDF document.", visible=True)}
+        except Exception as e:
+            print(f"Error occurred while writing the file: {e}")
+            return {error_box: gr.Textbox(value=f"Error occurred while writing the file: {e}", visible=True)}
+
+
+def respond(message, chat_history, 
+            language_choice, max_length, temperature,
+            num_return_sequences, top_p, no_repeat_ngram_size):
+        #No LLM here, just respond with a random pre-made message
+        if content == "":
+            bot_message = f"j'ai {max_length}" + random.choice(["Tell me more about it", 
+                                            "Cool, but I'm not interested", 
+                                            "Hmmmm, ok then"]) 
+        else:
+            bot_message = " j'ai besoin d'un modèle pour lire le {content[:3]}"
+        chat_history.append((message, bot_message))
+        return "", chat_history
+
+
+
+# Layout
+with gr.Blocks(theme=gr.themes.Soft()) as gradioApp:
+    with gr.Row():
+        with gr.Column(scale=1, min_width=100):
+            logo_gr = gr.Markdown(""" <img src="file/logo_neovision.png" alt="logo" style="width:400px;"/>""") 
+            # gr.Image("./logo_neovision.png")
+            about_gr = gr.Markdown(about)   
+            
+        with gr.Column(scale=2, min_width=500):
+            title1_gr= gr.Markdown(title1)
+            intro_gr = gr.Markdown(intro)
+
+            #  Upload several documents
+            content = ""
+            upload_button = gr.UploadButton("Browse files", label="Drag and drop your documents here", 
+                                            size="lg", scale=0, min_width=100,
+                                            file_types=["pdf"], file_count="multiple")
+            uploaded_check = gr.Radio(label="Uploaded documents", visible=False,
+                                      info="Do you want to use a supporting document?")
+            choose_btn = gr.Button(value="Choose", visible=False)
+            upload_button.upload(upload_file, upload_button, [uploaded_check, choose_btn])
+
+            # Read only one document
+            error_box = gr.Textbox(label="Reading files... ", visible=False)
+            choose_btn.click(read_content, inputs=uploaded_check, outputs=error_box)
+
+            # Select advanced options
+            gr.Markdown(""" ## Toolbox """)
+            with gr.Accordion(label="Select advanced options",open=False):
+                language_choice = gr.Dropdown(["English", "French"], label="Language", info="Choose your language")
+                max_length = gr.Slider(label="Token length", minimum=1, maximum=100, value=50, step=1)
+                temperature= gr.Slider(label="Temperature", minimum=0.1, maximum=1, value=0.8, step=0.1)
+                num_return_sequences= gr.Slider(label="Temperature", minimum=0.1, maximum=50, value=1, step=0.1)
+                top_p= gr.Slider(label="Temperature", minimum=0.1, maximum=1, value=0.8, step=0.1)
+                no_repeat_ngram_size= gr.Slider(label="Temperature", minimum=0.1, maximum=1, value=3, step=0.1)
+
+
+        # Chat
+        with gr.Column(scale=2, min_width=600):
+            title2_gr = gr.Markdown(title2)
+            chatbot = gr.Chatbot(label="Bot", height=500) 
+            msg = gr.Textbox(label="User", placeholder="Ask any question.")
+            chatbot_btn = gr.Button("Submit")
+            chatbot_btn.click(respond, inputs=[msg, chatbot, 
+                                               language_choice, max_length, temperature,
+                                               num_return_sequences, top_p, no_repeat_ngram_size],
+                                               outputs=[msg, chatbot])
+            clear = gr.ClearButton(components=[msg, chatbot], value="Clear console")
+
+
+gr.close_all()
+gradioApp.launch(share=True, enable_queue=True)
+
+

streamlit_app.py

@@ -0,0 +1,170 @@
+
+import streamlit as st
+
+# Set a custom background
+import torch
+from langchain import HuggingFacePipeline
+from langchain.chains import LLMChain, RetrievalQA
+from langchain.document_loaders import (
+    DirectoryLoader,
+    PyPDFLoader,
+    TextLoader,
+    UnstructuredPDFLoader,
+)
+from langchain.embeddings import HuggingFaceEmbeddings, LlamaCppEmbeddings
+from langchain.llms import LlamaCpp
+from langchain.prompts import PromptTemplate
+from langchain.text_splitter import (
+    CharacterTextSplitter,
+    RecursiveCharacterTextSplitter,
+)
+from langchain.vectorstores import Chroma
+from PIL import Image
+from streamlit_extras.add_vertical_space import add_vertical_space
+
+st.set_page_config(page_title="Welcome to our AI Question Answering Bot")
+
+with st.sidebar:
+    st.title('🤗💬 QA App')
+    st.markdown('''
+    ## About
+    This app is an LLM-powered chatbot built using:
+    - [Streamlit](<https://streamlit.io/>)
+    - [HugChat](<https://github.com/Soulter/hugging-chat-api>)
+    - Chat Model = llama2-chat-hf 7B 
+    - Retreiver model = all-MiniLM-L6-v2
+    
+    💡 Note: No API key required!
+    ''')
+    add_vertical_space(5)
+    st.write('Made with ❤️ by us')
+
+# logo = Image.open('logo.png')
+# st.image(logo, use_column_width=True)
+
+
+# Introduction
+st.markdown("""
+Welcome! This is not just any bot, it's a special one equipped with state-of-the-art natural language processing capabilities, and ready to answer your queries.
+
+
+Ready to explore? Let's get started!
+
+* Step 1: Upload a PDF document.
+* Step 2: Type in a question related to your document's content.
+* Step 3: Get your answer!
+
+Push clear cache before uploading a new doc !
+
+
+""")
+
+
+def write_text_file(content, file_path):
+    try:
+        with open(file_path, 'wb') as file:
+            file.write(content)
+        return True
+    except Exception as e:
+        print(f"Error occurred while writing the file: {e}")
+        return False
+
+
+# Wrap prompt template in a PromptTemplate object
+
+def set_qa_prompt():
+    # set prompt template
+    prompt_template = """<s>[INST] <<SYS>> Use the following pieces of context closed between $ to answer the question closed between |. If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
+    ${context}$ <</SYS>>
+    Question: |{question}|
+    Answer:[/INST]</s>"""
+    prompt = PromptTemplate(
+        template=prompt_template, input_variables=["context", "question"]
+    )
+    return prompt
+
+
+# Build RetrievalQA object
+
+def build_retrieval_qa(_llm, _prompt, _vectorstore):
+    dbqa = RetrievalQA.from_chain_type(llm=_llm,
+                                       chain_type='stuff',
+                                       retriever=_vectorstore.as_retriever(search_kwargs={'k': 3}),
+                                       return_source_documents=True,
+                                       chain_type_kwargs={'prompt': _prompt})
+    return dbqa
+
+
+# Instantiate QA object
+# @st.cache(allow_output_mutation=True)
+# @st.cache_resource()
+@st.cache(allow_output_mutation=True)
+def setup_dbqa(_texts):
+    print("setup_dbqa ...")
+    llm = HuggingFacePipeline.from_model_id(
+        model_id="NousResearch/Llama-2-13b-chat-hf",
+        task="text-generation",
+        model_kwargs={
+            "max_length": 1500, "load_in_8bit": True},
+    )
+
+    embeddings = HuggingFaceEmbeddings(model_name='sentence-transformers/all-MiniLM-L6-v2',
+                                       model_kwargs={'device': 'cpu'})
+
+    vectorstore = Chroma.from_documents(texts, embeddings, persist_directory="vectorstore")
+
+    prompt = set_qa_prompt()
+
+    return build_retrieval_qa(llm, prompt, vectorstore)
+
+
+def load_docs(uploaded_file):
+    print("loading docs ...")
+    content = uploaded_file.read()
+    file_path_aux = "./temp/file.pdf"
+    write_text_file(content, file_path_aux)
+    file_path = "./temp/"
+
+    loader = DirectoryLoader(file_path,
+                             glob="*.pdf",
+                             loader_cls=UnstructuredPDFLoader)
+    documents = loader.load()
+
+    # Split text from PDF into chunks
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1000,
+                                                   chunk_overlap=0,
+                                                   length_function=len,)
+    texts = text_splitter.split_documents(documents)
+    return texts
+
+
+# Set the background image
+# Load a PDF file
+uploaded_file = st.file_uploader("Choose a PDF file", type="pdf")
+
+if uploaded_file is not None:
+    st.write('Loading file')
+
+    texts = load_docs(uploaded_file)
+    model = setup_dbqa(texts)
+
+    # Build and persist FAISS vector store
+
+    question = st.text_input('Ask a question:')
+
+    if question:
+        # Placeholder for chatbot logic to generate an answer based on the question and the PDF content
+
+        answer = model({'query': question})
+        # The below is just a hardcoded response
+        print(question)
+        print(answer)
+
+        # st.write('Question: ', answer["query"])
+        st.write('Question: ', answer["query"])
+        st.write('Answer: ', answer["result"])
+        st.write('Source documents: ', answer["source_documents"])
+
+# if st.button("Clear cache before loading new document"):
+#     # Clears all st.cache_resource caches:
+#     st.cache_resource.clear()

test_gradio.py

@@ -0,0 +1,188 @@
+
+# in folder ==> gradio test_gradio.py ( it won’t provide the automatic reload mechanism?)
+
+import gradio as gr
+import numpy as np
+import random
+import time
+
+# def greet(name):
+#     return "Hello " + name + "!"
+
+# demo = gr.Interface(fn=greet, inputs=gr.Textbox(lines=2, placeholder="Name Here..."), outputs="text")
+    
+# demo.launch()   
+
+
+# def greet2(name, is_morning, temperature):
+#     salutation = "Good morning" if is_morning else "Good evening"
+#     greeting = f"{salutation} {name}. It is {temperature} degrees today"
+#     celsius = (temperature - 32) * 5 / 9
+#     return greeting, round(celsius, 2)
+
+# demo = gr.Interface(
+#     fn=greet2,
+#     inputs=["text", "checkbox", gr.Slider(0, 100)],
+#     outputs=["text", "number"],
+# )
+# demo.launch()
+
+
+# def sepia(input_img):
+#     sepia_filter = np.array([
+#         [0.393, 0.769, 0.189], 
+#         [0.349, 0.686, 0.168], 
+#         [0.272, 0.534, 0.131]
+#     ])
+#     sepia_img = input_img.dot(sepia_filter.T)
+#     sepia_img /= sepia_img.max()
+#     return sepia_img
+
+# demo = gr.Interface(sepia, gr.Image(shape=(200, 200)), "image")
+# demo.launch()
+
+
+
+
+
+def yes_man(message, history):
+    if message.endswith("?"):
+        return "Yes"
+    else:
+        return "Ask me anything!"
+
+
+
+# gr.ChatInterface(
+#     yes_man,
+#     chatbot=gr.Chatbot(height=300),
+#     textbox=gr.Textbox(placeholder="Ask a question about the uploaded PDF document.", container=False, scale=7),
+#     title="Gradio QA Bot",
+#     description=f"{intro}",
+#     theme="soft",
+#     examples=["What is the title of the document?", "Summarize the main ideas of the documents"],
+#     cache_examples=True,
+#     retry_btn=None,
+#     undo_btn="Delete Previous",
+#     clear_btn="Clear",
+# ).launch()
+
+
+# intro = "Welcome! This is not just any bot, ..."
+title1 = "QA App"
+title2 = "Gradio QA Bot"
+
+
+def file_upload(input_file):
+    # Process the uploaded file
+    if input_file is not None:
+        # Save the uploaded file or perform any desired operations
+        file_path = "/tmp/file.pdf"
+        content = input_file.read()
+        try:
+            with open(file_path, 'wb') as file:
+                file.write(content)
+            return [f"File '{input_file.name}' uploaded successfully in {file_path}.",file_path] 
+        except Exception as e:
+            return f"Error occurred while writing the file: {e}"
+    return ["No file uploaded.", file_path] 
+
+
+def crash(test, file):
+    return("ok")
+
+
+
+gr.ChatInterface(
+        yes_man,
+        chatbot=gr.Chatbot(height=300),
+        textbox=gr.Textbox(placeholder="Ask a question about the uploaded PDF document.", container=False, scale=7),
+        title="Gradio QA Bot",
+        description="blabla",
+        theme="soft",
+        examples=["What is the title of the document?", "Summarize the main ideas of the documents"],
+        cache_examples=True,
+        retry_btn=None,
+        undo_btn="Delete Previous",
+        clear_btn="Clear",
+    ).launch()
+
+
+# with gr.Blocks() as demo:
+#     intro = gr.Markdown("""Welcome! This is not just any bot, it's a special one equipped with state-of-the-art natural language processing capabilities, and ready to answer your queries.
+
+#         Ready to explore? Let's get started!
+
+#         * Step 1: Upload a PDF document.
+#         * Step 2: Type in a question related to your document's content.
+#         * Step 3: Get your answer!
+
+#         Push clear cache before uploading a new doc!
+#         """)
+    
+#     # Create a Gradio interface with a file upload input
+#     iface = gr.Interface(
+#         fn=file_upload,
+#         inputs=gr.File(),
+#         outputs=["text", gr.File()],
+#         title=title1,
+#         description="Drag and drop your document here")
+    
+
+#     # bot = gr.Interface(crash,
+#     #             inputs=[gr.Textbox(lines=2, placeholder="Ask a question about the uploaded PDF document."), gr.File()],
+#     #             outputs=[gr.Chatbot(height=300)],
+#     #             title="Gradio QA Bot",
+#     #             description=f"{intro}",
+#     #             theme="soft",
+#     #             examples=["What is the title of the document?", "Summarize the main ideas of the documents"],
+#     #             cache_examples=True,
+#     #             retry_btn=None,
+#     #             undo_btn="Delete Previous",
+#     #             clear_btn="Clear")
+    
+#     # gr.ChatInterface(
+#     #     yes_man,
+#     #     chatbot=gr.Chatbot(height=300),
+#     #     textbox=gr.Textbox(placeholder="Ask a question about the uploaded PDF document.", container=False, scale=7),
+#     #     title="Gradio QA Bot",
+#     #     description=f"{intro}",
+#     #     theme="soft",
+#     #     examples=["What is the title of the document?", "Summarize the main ideas of the documents"],
+#     #     cache_examples=True,
+#     #     retry_btn=None,
+#     #     undo_btn="Delete Previous",
+#     #     clear_btn="Clear",
+#     # )
+
+# demo.launch()
+
+
+
+# bot
+iface = gr.Interface(qa_bot, 
+                     inputs=["file", gr.Textbox(placeholder="Ask a question about the uploaded PDF document.", container=False, scale=7)],
+                     outputs="text",
+                     title=title2,
+                     description="Ask a question about the uploaded PDF document.",
+                     theme="soft",
+                    examples=["What is the title of the document?", "Summarize the main ideas of the documents"],
+                    cache_examples=True,
+                    retry_btn=None,
+                    undo_btn="Delete Previous",
+                    clear_btn="Clear")
+#### OR 
+
+iface =  gr.ChatInterface(
+            qa_bot,
+            chatbot=gr.Chatbot(height=300),
+            textbox=gr.Textbox(placeholder="Ask a question about the uploaded PDF document.", container=False, scale=7),
+            title=title2,
+            description="Ask a question about the uploaded PDF document.",
+            theme="soft",
+            examples=["What is the title of the document?", "Summarize the main ideas of the documents"],
+            cache_examples=True,
+            retry_btn=None,
+            undo_btn="Delete Previous",
+            clear_btn="Clear",
+        )

test_streamlit.py

@@ -0,0 +1,10 @@
+import streamlit as st
+import pandas as pd
+    
+st.write("""
+# My first app
+Hello *world!*
+""")
+    
+df = pd.read_csv("my_data.csv")
+st.line_chart(df)