Upload folder using huggingface_hub

Dockerfile

@@ -0,0 +1,24 @@
+FROM python:3.10 
+
+WORKDIR /usr/src/app 
+# RUN mkdir /usr/src/app/greenprocurementdb2
+# RUN mkdir /usr/src/app/pickle
+# RUN mkdir /usr/src/app/data
+
+
+ 
+COPY requirements.txt ./ 
+COPY ./greenprocurementdb2 ./greenprocurementdb2
+COPY ./pickle ./pickle
+COPY ./data ./data
+COPY ./model ./model
+COPY ./summaries ./summaries
+
+RUN pip install --upgrade pip
+RUN pip install --no-cache-dir -r requirements.txt  
+
+COPY ./app.py ./
+
+EXPOSE 7860
+
+CMD [ "python", "./app.py" ] 

README.md

@@ -1,12 +1,6 @@
 ---
-title: Green Procurement
-emoji: 🦀
-colorFrom: yellow
-colorTo: purple
-sdk: gradio
-sdk_version: 3.44.4
+title: Green_Procurement
 app_file: app.py
-pinned: false
+sdk: gradio
+sdk_version: 3.44.3
 ---
-
-Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,98 @@
+import os
+
+import gradio as gr
+import openai
+from pathlib import Path
+
+from langchain import PromptTemplate, LLMChain
+import qdrant_client
+from dotenv import load_dotenv
+# from langchain.chains.qa_with_sources import load_qa_with_sources_chain
+from langchain.chains import RetrievalQA
+from langchain.chat_models import ChatOpenAI
+from langchain.chains.question_answering import load_qa_chain
+from langchain.embeddings.openai import OpenAIEmbeddings
+from langchain.llms import AzureOpenAI, OpenAI
+from langchain.text_splitter import CharacterTextSplitter
+from langchain.vectorstores import Qdrant
+from qdrant_client import QdrantClient
+
+# Constants
+collection_name="10ks"
+# collection_name="collectiveagreements"
+
+
+# Load the environment variables with the Azure OpenAI API key
+load_dotenv()
+
+# Initialize Azure OpenAI
+# openai.api_type = os.getenv("OPENAI_API_TYPE")
+# openai.api_base = os.getenv("OPENAI_API_BASE")
+openai.api_key = os.getenv("OPENAI_API_KEY")
+# openai.api_version = os.getenv("OPENAI_API_VERSION")
+
+# The data was vectorized with ADA, so we'll use that to convert our
+# query into a vector
+embeddings = OpenAIEmbeddings(model="text-embedding-ada-002") #, chunk_size=1)
+
+dbclient = QdrantClient("localhost", port=6333, grpc_port=6334, prefer_grpc=True)
+
+index=Qdrant(client=dbclient, collection_name=collection_name, embeddings=embeddings, vector_name="fragmentvector")
+
+# # Load the FAISS index
+# index = dbclient.retrieve. .load_local(
+#     "collectiveagreements.db",
+#     OpenAIEmbeddings(chunk_size=1, model="text-embedding-ada-002"),
+# )
+
+# Open a connection to render the search results into test- this uses davinci-002.
+llm = OpenAI(deployment_name="davinci", temperature=0)
+
+# Open op a connection to do the querying
+# Chain type can be stuff, map_reduce or refine
+# chain = load_qa_with_sources_chain(llm, chain_type="map_reduce")
+
+def docquery(question):
+    docs = index.similarity_search(question)
+    print("Length of answer: ", len(docs))
+    # Process the query and return the results
+    llm = ChatOpenAI(model_name="gpt-3.5-turbo", temperature=0)
+    qa_chain = RetrievalQA.from_chain_type(llm,retriever=index.as_retriever(), return_source_documents=True)
+    output=qa_chain({"query": question})
+    # output = chain.run(input_documents=docs, question=question)
+    print("Is the error here?", output  )
+    # myanswer = "##" + output.split("SOURCES")[0]
+    references = ""
+    # print("Docs:", docs)
+    for i in docs:
+        print("item: ", i.page_content)
+        references = (
+            references
+            + "**"
+            + "** \n"
+            + i.page_content.replace("\n", "")
+            + "\n\n"
+        )
+    return output['result'], references
+
+with gr.Blocks(title="Collective Agreement Search") as blocks:
+    appname = gr.Markdown(value="# 10K filings search")
+    appdesc = gr.Markdown(
+        value="## The tabs below demonstration different ways to query the data."
+    )
+
+    with gr.Tab("Ask a question"):
+        appdesc = gr.Markdown(
+            value="### This is a demo of an OpenAI-based question answering system. Type in a question and the system will return the answer and the source document."
+        )
+        question = gr.Textbox(
+            lines=1,
+            label="Question: press enter to submit",
+            value="Where is Babcock's head office?",
+        )
+        answer = gr.Markdown(label="Answer")
+        references = gr.Markdown(label="References")
+        question.submit(docquery, question, outputs=[answer, references])
+
+
+blocks.launch(share=True, server_name="0.0.0.0", server_port=8080)