first commit

.gitattributes

@@ -0,0 +1,3 @@
+*.sqlite3 filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,12 @@
+db
+db_bgai
+db2
+models
+source_documents
+app_v2.py
+app_v3.py
+app.py
+ingest.py
+ingest_v3.py
+.env
+README.md

app_palm.py

@@ -0,0 +1,94 @@
+from langchain import PromptTemplate, LLMChain
+from langchain.llms import CTransformers, HuggingFacePipeline, GooglePalm
+import os
+from langchain.text_splitter import RecursiveCharacterTextSplitter
+from langchain.vectorstores import Chroma
+from langchain.chains import RetrievalQA
+from langchain.embeddings import HuggingFaceBgeEmbeddings, HuggingFaceEmbeddings
+from io import BytesIO
+from langchain.document_loaders import PyPDFLoader
+import gradio as gr
+import chromadb
+from dotenv import load_dotenv
+from constants import CHROMA_SETTINGS
+from io import BytesIO
+import gradio as gr
+import torch
+from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline, AutoModelForSeq2SeqLM, AutoModel
+import gc
+
+from langchain.schema.runnable import RunnableLambda, RunnablePassthrough
+from langchain.chat_models import ChatGooglePalm
+import google.generativeai as genai
+from langchain_google_genai import GoogleGenerativeAIEmbeddings
+
+gc.collect()
+torch.cuda.empty_cache()
+
+device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')
+
+
+#model= AutoModelForCausalLM.from_pretrained(local_llm, device_map= device)
+llm= ChatGooglePalm()
+#llm= HuggingFacePipeline.from_model_id(model_id=local_llm, task='text-generation', device=0, pipeline_kwargs={"max_new_tokens": 1000})
+
+embeddings_model_name = os.environ.get('EMBEDDINGS_MODEL_NAME')
+persist_directory = os.environ.get('PERSIST_DIRECTORY')
+target_source_chunks = int(os.environ.get('TARGET_SOURCE_CHUNKS',4))
+google_api_key= os.environ.get('GOOGLE_API_KEY')
+if not load_dotenv():
+    print("Could not load .env file or it is empty. Please check if it exists and is readable.")
+    exit(1)
+    
+print("Loading embeddings model...")
+embeddings = GoogleGenerativeAIEmbeddings(model="models/embedding-001")
+#embeddings= pipeline("feature-extraction", model="WhereIsAI/UAE-Large-V1")
+# Chroma client
+chroma_client = chromadb.PersistentClient(settings=CHROMA_SETTINGS , path=persist_directory)
+db = Chroma(persist_directory=persist_directory, embedding_function=embeddings, client_settings=CHROMA_SETTINGS, client=chroma_client)
+
+prompt_template = """Use the following pieces of information to answer the user's question.
+If you don't know the answer, just say that you don't know, don't try to make up an answer.
+
+Context: {context}
+Question: {question}
+
+Only return the helpful answer below and nothing else.
+Helpful answer:
+"""
+
+prompt = PromptTemplate(template=prompt_template, input_variables=['context', 'question'])
+retriever = db.as_retriever(search_kwargs={"k": target_source_chunks})
+# activate/deactivate the streaming StdOut callback for LLMs
+
+chain_type_kwargs = {"prompt": prompt}
+
+input_gradio= gr.Text(
+    label="Prompt",
+    show_label=False,
+    max_lines=2,
+    placeholder="Enter your question here",
+    container=False,
+    
+)
+
+
+def get_response(input_gradio ):
+    query=input_gradio
+    qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever, return_source_documents= False, chain_type_kwargs=chain_type_kwargs, verbose=True)
+    response= qa(query)
+    return response['result']
+
+iface= gr.Interface(
+    fn=get_response,
+    inputs=input_gradio,
+    outputs="text",
+    title="Tsetlin Machine Chatbot",
+    description="A chatbot that uses the LLM to answer anything regarding TM",
+    allow_flagging='never'
+    
+)
+    # Interactive questions and answers
+iface.launch()
+
+

constants.py

@@ -0,0 +1,14 @@
+import os
+from dotenv import load_dotenv
+from chromadb.config import Settings
+
+load_dotenv()
+
+PERSIST_DIRECTORY = os.environ.get('PERSIST_DIRECTORY')
+if PERSIST_DIRECTORY is None:
+    raise Exception("please put the directory path in chroma db")
+
+CHROMA_SETTINGS = Settings(
+        persist_directory=PERSIST_DIRECTORY,
+        anonymized_telemetry=False
+)

db_google/chroma.sqlite3

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8eba354226ba88c591f1c149f3e965403e48dba1f085f71c7713bae1bd895f1f
+size 39329792

db_google/e072cde8-61bc-4eaa-9876-6068c95e6f74/data_level0.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0b2cb81ecbebf24556e4af8b17c5f6606e0b4a4b428c64814f74c798c38ec50e
+size 16060000

db_google/e072cde8-61bc-4eaa-9876-6068c95e6f74/header.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cd215c9f6b14de8a55607f23f90af168c5493da60c02482ee8f7087809be863
+size 100

db_google/e072cde8-61bc-4eaa-9876-6068c95e6f74/index_metadata.pickle

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2dc24cfc543dfc34ed403220c5ffa07e3d2caf5143d6eb2cb731d1dcd46d6336
+size 288034

db_google/e072cde8-61bc-4eaa-9876-6068c95e6f74/length.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ad4eb880e8012f662c48d0a8492c6006672085d2e3345ab37d4c05076c9a747
+size 20000

db_google/e072cde8-61bc-4eaa-9876-6068c95e6f74/link_lists.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b22486b1e0558438145eb5c5c3c063e3ec5f38af73a3a04639d53d13ea0b50a
+size 44752

requirements.txt

@@ -0,0 +1,17 @@
+langchain==0.0.274
+gradio==3.50.2
+gpt4all==1.0.8
+chromadb==0.4.7
+urllib3==2.0.4
+PyMuPDF==1.23.1
+python-dotenv==1.0.0
+unstructured==0.10.8
+extract-msg==0.45.0
+tabulate==0.9.0
+pandoc==2.3
+pypandoc==1.11
+tqdm==4.66.1
+sentence_transformers==2.2.2
+pypdf
+google-generativeai
+protobuf==3.20.*