Upload 17 files

.gitattributes

@@ -33,3 +33,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+Ehlers-Danlos-1/2024_EDS_2.pdf filter=lfs diff=lfs merge=lfs -text
+Ehlers-Danlos-1/2024_EDS_3.pdf filter=lfs diff=lfs merge=lfs -text
+Ehlers-Danlos-1/2024_EDS_4.pdf filter=lfs diff=lfs merge=lfs -text
+Ehlers-Danlos-1/2024_EDS_5.pdf filter=lfs diff=lfs merge=lfs -text
+Ehlers-Danlos-1/Unknown_EDS_1.pdf filter=lfs diff=lfs merge=lfs -text
+Ehlers-Danlos-1/Unknown_EDS_5.pdf filter=lfs diff=lfs merge=lfs -text

Ehlers-Danlos-1/2024_EDS_2.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46fc736ff4174473e0a846b7ca8430c140d89cd2c9f663e105bc48b33f8d9c99
+size 2616000

Ehlers-Danlos-1/2024_EDS_3.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7fef5c8c375297158ad7ad63166405ca7ce4ac511371a8454fe9df972755b0fe
+size 10344738

Ehlers-Danlos-1/2024_EDS_4.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:25db35c77fd6aeba6b15278671a462b30ffbb6f97eb5f221e0459f6d11c0f8ed
+size 1071576

Ehlers-Danlos-1/2024_EDS_5.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:57ef98bcb445da6abda66de35204634bd81d8c6dcdf53bfc3be54447ec9ad0ad
+size 2772421

Ehlers-Danlos-1/Unknown_EDS_1.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dbeaf13d3298a00bc1c7acfba3177a0c639f677e0f0941452709fe60542052d4
+size 21553835

Ehlers-Danlos-1/Unknown_EDS_5.pdf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c5a77524b6bb4dca40798af5ff3e3c622216a13ac21a60d9befce255977b47a
+size 1847313

app.py

@@ -0,0 +1,104 @@
+import os
+import glob
+from pathlib import Path
+import gradio as gr
+import nest_asyncio
+import dill as pickle
+import streamlit as st
+
+# Ensure async compatibility in Jupyter
+nest_asyncio.apply()
+
+# Import OpenAI key with helper function
+from helper import get_openai_api_key
+OPENAI_API_KEY = get_openai_api_key()
+
+# Define the path to the directory containing the PDF files
+folder_path = 'Ehlers-Danlos-1'
+
+# Get the list of all PDF files in the directory
+pdf_files = glob.glob(os.path.join(folder_path, '*.pdf'))
+print(pdf_files)
+
+# Extract just the filenames (optional)
+pdf_filenames = [os.path.basename(pdf) for pdf in pdf_files]
+print(pdf_filenames)
+
+# Import utilities
+from utils import get_doc_tools
+
+# Truncate function names if necessary
+def truncate_function_name(name, max_length=64):
+    return name if len(name) <= max_length else name[:max_length]
+
+# Path to save/load serialized tools
+tools_cache_path = 'tools_cache.pkl'
+
+# Initialize paper_to_tools_dict
+paper_to_tools_dict = {}
+
+# Check if the cache file exists and is not empty
+if os.path.exists(tools_cache_path) and os.path.getsize(tools_cache_path) > 0:
+    try:
+        with open(tools_cache_path, 'rb') as f:
+            paper_to_tools_dict = pickle.load(f)
+    except EOFError:
+        print("Cache file is corrupted. Recreating tools.")
+        paper_to_tools_dict = {}
+else:
+    print("Cache file does not exist or is empty. Recreating tools.")
+
+# Create tools for each PDF if not loaded from cache
+if not paper_to_tools_dict:
+    for pdf in pdf_files:
+        print(f"Getting tools for paper: {pdf}")
+        vector_tool, summary_tool = get_doc_tools(pdf, Path(pdf).stem)
+        paper_to_tools_dict[pdf] = [vector_tool, summary_tool]
+    
+    # Save tools to cache
+    with open(tools_cache_path, 'wb') as f:
+        pickle.dump(paper_to_tools_dict, f)
+
+
+# Combine all tools into a single list
+all_tools = [t for pdf in pdf_files for t in paper_to_tools_dict[pdf]]
+
+# Define an object index and retriever over these tools
+from llama_index.core import VectorStoreIndex
+from llama_index.core.objects import ObjectIndex
+
+obj_index = ObjectIndex.from_objects(
+    all_tools,
+    index_cls=VectorStoreIndex,
+)
+
+obj_retriever = obj_index.as_retriever(similarity_top_k=3)
+
+# Initialize the OpenAI LLM
+from llama_index.llms.openai import OpenAI
+llm = OpenAI(model="gpt-3.5-turbo")
+
+# Set up the agent
+from llama_index.core.agent import FunctionCallingAgentWorker
+from llama_index.core.agent import AgentRunner
+
+agent_worker = FunctionCallingAgentWorker.from_tools(
+    tool_retriever=obj_retriever,
+    llm=llm, 
+    verbose=True
+)
+agent = AgentRunner(agent_worker)
+
+# Define the function to query the agent
+def ask_agent(question):
+    response = agent.query(question)
+    return str(response)
+
+# Streamlit interface
+st.title("EDS Research Agent")
+
+question = st.text_input("Ask a question:")
+if question:
+    answer = ask_agent(question)
+    st.write(answer)
+

helper.py

@@ -0,0 +1,13 @@
+# Add your utilities or helper functions to this file.
+
+import os
+from dotenv import load_dotenv, find_dotenv
+
+# these expect to find a .env file at the directory above the lesson.                                                                                                                     # the format for that file is (without the comment)                                                                                                                                       #API_KEYNAME=AStringThatIsTheLongAPIKeyFromSomeService                                                                                                                                     
+def load_env():
+    _ = load_dotenv(find_dotenv())
+
+def get_openai_api_key():
+    load_env()
+    openai_api_key = os.getenv("OPENAI_API_KEY")
+    return openai_api_key

ragas_eval.py

@@ -0,0 +1,36 @@
+import os
+import sys
+from helper import get_openai_api_key
+
+venv_path = os.path.join(os.path.dirname(__file__), 'venv', 'Lib', 'python3.12', 'site-packages')
+sys.path.append(venv_path)
+
+os.environ["OPENAI_API_KEY"] = get_openai_api_key()
+
+from langchain_community.document_loaders import DirectoryLoader
+loader = DirectoryLoader("Ehlers-Danlos-1")
+documents = loader.load()
+
+for document in documents:
+    document.metadata['filename'] = document.metadata['source']
+
+from ragas.testset.generator import TestsetGenerator
+from ragas.testset.evolutions import simple, reasoning, multi_context
+from langchain_openai import ChatOpenAI, OpenAIEmbeddings
+
+# generator with openai models
+generator_llm = ChatOpenAI(model="gpt-3.5-turbo")
+critic_llm = ChatOpenAI(model="gpt-4")
+embeddings = OpenAIEmbeddings()
+
+generator = TestsetGenerator.from_langchain(
+    generator_llm,
+    critic_llm,
+    embeddings
+)
+
+# generate testset
+testset = generator.generate_with_langchain_docs(documents, test_size=10, distributions={simple: 0.5, reasoning: 0.25, multi_context: 0.25})
+print("!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!")
+print(testset)
+testset.to_pandas()

test.py

@@ -0,0 +1,26 @@
+import sys
+import os
+
+# Add the virtual environment's site-packages to sys.path
+# Replace 'pythonX.Y' with your Python version, e.g., 'python3.8'
+venv_path = os.path.join(os.path.dirname(__file__), 'venv', 'lib', 'site-packages')
+sys.path.append(venv_path)
+
+# Ensure the directory structure is recognized as a package
+# You can verify by listing the contents of the directory
+print("sys.path:", sys.path)
+print("Contents of venv_path:", os.listdir(venv_path))
+
+# Now import the TestsetGenerator
+try:
+    from ragas.testset.generator import TestsetGenerator
+    print("Successfully imported TestsetGenerator.")
+except ImportError as e:
+    print("ImportError:", e)
+
+# Use the imported function or class
+try:
+    generator = TestsetGenerator()
+    print("Successfully created a TestsetGenerator instance.")
+except Exception as e:
+    print("Error creating TestsetGenerator instance:", e)

utils.py

@@ -0,0 +1,82 @@
+from llama_index.core import SimpleDirectoryReader
+from llama_index.core.node_parser import SentenceSplitter
+from llama_index.core import Settings
+from llama_index.llms.openai import OpenAI
+from llama_index.embeddings.openai import OpenAIEmbedding
+from llama_index.core import SummaryIndex, VectorStoreIndex
+from llama_index.core.tools import QueryEngineTool
+from llama_index.core.query_engine.router_query_engine import RouterQueryEngine
+from llama_index.core.selectors import LLMSingleSelector
+
+from llama_index.core import SimpleDirectoryReader, VectorStoreIndex, SummaryIndex
+from llama_index.core.node_parser import SentenceSplitter
+from llama_index.core.tools import FunctionTool, QueryEngineTool
+from llama_index.core.vector_stores import MetadataFilters, FilterCondition
+from typing import List, Optional
+
+
+
+def get_doc_tools(
+    file_path: str,
+    name: str,
+) -> str:
+    """Get vector query and summary query tools from a document."""
+
+    # load documents
+    documents = SimpleDirectoryReader(input_files=[file_path]).load_data()
+    splitter = SentenceSplitter(chunk_size=1024)
+    nodes = splitter.get_nodes_from_documents(documents)
+    vector_index = VectorStoreIndex(nodes)
+    
+    def vector_query(
+        query: str, 
+        page_numbers: Optional[List[str]] = None
+    ) -> str:
+        """Use to answer questions over a given paper.
+    
+        Useful if you have specific questions over the paper.
+        Always leave page_numbers as None UNLESS there is a specific page you want to search for.
+    
+        Args:
+            query (str): the string query to be embedded.
+            page_numbers (Optional[List[str]]): Filter by set of pages. Leave as NONE 
+                if we want to perform a vector search
+                over all pages. Otherwise, filter by the set of specified pages.
+        
+        """
+    
+        page_numbers = page_numbers or []
+        metadata_dicts = [
+            {"key": "page_label", "value": p} for p in page_numbers
+        ]
+        
+        query_engine = vector_index.as_query_engine(
+            similarity_top_k=2,
+            filters=MetadataFilters.from_dicts(
+                metadata_dicts,
+                condition=FilterCondition.OR
+            )
+        )
+        response = query_engine.query(query)
+        return response
+        
+    
+    vector_query_tool = FunctionTool.from_defaults(
+        name=f"vector_tool_{name}",
+        fn=vector_query
+    )
+    
+    summary_index = SummaryIndex(nodes)
+    summary_query_engine = summary_index.as_query_engine(
+        response_mode="tree_summarize",
+        use_async=True,
+    )
+    summary_tool = QueryEngineTool.from_defaults(
+        name=f"summary_tool_{name}",
+        query_engine=summary_query_engine,
+        description=(
+            f"Useful for summarization questions related to {name}"
+        ),
+    )
+
+    return vector_query_tool, summary_tool