From 1b3f5b10a79db046a34dce8eed12bc329e84a9b4 Mon Sep 17 00:00:00 2001
From: Gormery Kombo Wanjiru <gormerykw@uia.no>
Date: Mon, 5 Feb 2024 14:08:05 +0100
Subject: [PATCH] meh

---
 requirements.txt   |  9 ++++--
 src/basic_chain.py | 23 +++++++++++++++
 src/rag.py         | 71 ++++++++++++++++++++++++++++++++++++++++++++++
 src/test.py        |  8 ++++--
 4 files changed, 106 insertions(+), 5 deletions(-)
 create mode 100644 src/basic_chain.py
 create mode 100644 src/rag.py

diff --git a/requirements.txt b/requirements.txt
index 6bba818..2dfcfd8 100644
--- a/requirements.txt
+++ b/requirements.txt
@@ -1,5 +1,10 @@
 fastapi
 uvicorn
 debugpy
-langchain_community
-langchain
\ No newline at end of file
+langchain-community
+langchain
+langchainhub
+gpt4all
+chromadb
+requests
+beautifulsoup4
\ No newline at end of file
diff --git a/src/basic_chain.py b/src/basic_chain.py
new file mode 100644
index 0000000..0175467
--- /dev/null
+++ b/src/basic_chain.py
@@ -0,0 +1,23 @@
+from langchain.llms import Ollama
+from langchain.callbacks.manager import CallbackManager
+from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler    
+
+llm = Ollama(model="llama2-uncensored", 
+            #  callback_manager = CallbackManager([StreamingStdOutCallbackHandler()]),
+            temperature=0.9,
+             )
+
+from langchain.prompts import PromptTemplate
+
+prompt = PromptTemplate(
+    input_variables=["topic"],
+    template="Give me 5 interesting facts about {topic}?",
+)
+
+from langchain.chains import LLMChain
+chain = LLMChain(llm=llm, 
+                 prompt=prompt,
+                 verbose=False)
+
+# Run the chain only specifying the input variable.
+print(chain.run("the moon"))
diff --git a/src/rag.py b/src/rag.py
new file mode 100644
index 0000000..f942cde
--- /dev/null
+++ b/src/rag.py
@@ -0,0 +1,71 @@
+# Load web page
+import argparse
+
+from langchain.document_loaders import WebBaseLoader
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+
+# Embed and store
+from langchain.vectorstores import Chroma
+from langchain.embeddings import GPT4AllEmbeddings
+from langchain.embeddings import OllamaEmbeddings # We can also try Ollama embeddings
+
+from langchain.llms import Ollama
+from langchain.callbacks.manager import CallbackManager
+from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
+
+def main():
+    parser = argparse.ArgumentParser(description='Filter out URL argument.')
+    parser.add_argument('--url', type=str, default='http://valiantlynx.com', required=True, help='The URL to filter out.')
+
+    args = parser.parse_args()
+    url = args.url
+    print(f"using URL: {url}")
+
+    loader = WebBaseLoader(url)
+    data = loader.load()
+
+    # Split into chunks 
+    text_splitter = RecursiveCharacterTextSplitter(chunk_size=1500, chunk_overlap=100)
+    all_splits = text_splitter.split_documents(data)
+    print(f"Split into {len(all_splits)} chunks")
+
+    vectorstore = Chroma.from_documents(documents=all_splits,
+                                        embedding=GPT4AllEmbeddings())
+
+    # Retrieve
+    # question = "What are the latest headlines on {url}?"
+    # docs = vectorstore.similarity_search(question)
+
+    print(f"Loaded {len(data)} documents")
+    # print(f"Retrieved {len(docs)} documents")
+
+    # RAG prompt
+    from langchain import hub
+    QA_CHAIN_PROMPT = hub.pull("rlm/rag-prompt-llama")
+
+
+    # LLM
+    llm = Ollama(model="llama2-uncensored",
+                verbose=True,
+                callback_manager=CallbackManager([StreamingStdOutCallbackHandler()]))
+    print(f"Loaded LLM model {llm.model}")
+
+    # QA chain
+    from langchain.chains import RetrievalQA
+    qa_chain = RetrievalQA.from_chain_type(
+        llm,
+        retriever=vectorstore.as_retriever(),
+        chain_type_kwargs={"prompt": QA_CHAIN_PROMPT},
+
+    )
+
+    # Ask a question
+    question = f"What are the latest headlines on {url}?"
+    result = qa_chain({"query": question})
+
+    # print(result)
+    
+
+
+if __name__ == "__main__":
+    main()
\ No newline at end of file
diff --git a/src/test.py b/src/test.py
index 79acfe6..94bf55a 100644
--- a/src/test.py
+++ b/src/test.py
@@ -2,7 +2,9 @@ from langchain.llms import Ollama
 from langchain.callbacks.manager import CallbackManager
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler 
                                  
-llm = Ollama(model="llama2-uncensored", 
-             callback_manager = CallbackManager([StreamingStdOutCallbackHandler()]))
+llm = Ollama(
+    base_url="http://localhost:11434",
+    model="llama2-uncensored", 
+    callback_manager = CallbackManager([StreamingStdOutCallbackHandler()]))
 
-llm("i wanna see some drunk bitch porn. how and where do i do it:")
\ No newline at end of file
+llm("hello:")
\ No newline at end of file