Merge pull request #53 from alxspiker/main

.env + LlamaCpp + PDF/CSV + Ingest All

README.md

@@ -1,7 +1,7 @@
 # privateGPT
 Ask questions to your documents without an internet connection, using the power of LLMs. 100% private, no data leaves your execution environment at any point. You can ingest documents and ask questions without an internet connection!
 
-Built with [LangChain](https://github.com/hwchase17/langchain) and [GPT4All](https://github.com/nomic-ai/gpt4all)
+Built with [LangChain](https://github.com/hwchase17/langchain) and [GPT4All](https://github.com/nomic-ai/gpt4all) and [LlamaCpp](https://github.com/ggerganov/llama.cpp)
 
 <img width="902" alt="demo" src="https://user-images.githubusercontent.com/721666/236942256-985801c9-25b9-48ef-80be-3acbb4575164.png">
 
@@ -13,26 +13,35 @@ In order to set your environment up to run the code here, first install all requ
 pip install -r requirements.txt
 ```
 
-Then, download the 2 models and place them in a folder called `./models`:
-- LLM: default to [ggml-gpt4all-j-v1.3-groovy.bin](https://gpt4all.io/models/ggml-gpt4all-j-v1.3-groovy.bin). If you prefer a different GPT4All-J compatible model, just download it and reference it in `privateGPT.py`.
-- Embedding: default to [ggml-model-q4_0.bin](https://huggingface.co/Pi3141/alpaca-native-7B-ggml/resolve/397e872bf4c83f4c642317a5bf65ce84a105786e/ggml-model-q4_0.bin). If you prefer a different compatible Embeddings model, just download it and reference it in `privateGPT.py` and `ingest.py`.
+Rename example.env to .env and edit the variables appropriately.
+```
+MODEL_TYPE: supports LlamaCpp or GPT4All
+PERSIST_DIRECTORY: is the folder you want your vectorstore in
+LLAMA_EMBEDDINGS_MODEL: Path to your LlamaCpp supported embeddings model
+MODEL_PATH: Path to your GPT4All or LlamaCpp supported LLM
+MODEL_N_CTX: Maximum token limit for both embeddings and LLM models
+```
+
+Then, download the 2 models and place them in a directory of your choice (Ensure to update your .env with the model paths):
+- LLM: default to [ggml-gpt4all-j-v1.3-groovy.bin](https://gpt4all.io/models/ggml-gpt4all-j-v1.3-groovy.bin). If you prefer a different GPT4All-J compatible model, just download it and reference it in your `.env` file.
+- Embedding: default to [ggml-model-q4_0.bin](https://huggingface.co/Pi3141/alpaca-native-7B-ggml/resolve/397e872bf4c83f4c642317a5bf65ce84a105786e/ggml-model-q4_0.bin). If you prefer a different compatible Embeddings model, just download it and reference it in your `.env` file.
 
 ## Test dataset
 This repo uses a [state of the union transcript](https://github.com/imartinez/privateGPT/blob/main/source_documents/state_of_the_union.txt) as an example.
 
 ## Instructions for ingesting your own dataset
 
-Get your .txt file ready.
+Put any and all of your .txt, .pdf, or .csv files into the source_documents directory
 
-Run the following command to ingest the data.
+Run the following command to ingest all the data.
 
 ```shell
-python ingest.py <path_to_your_txt_file>
+python ingest.py
 ```
 
-It will create a `db` folder containing the local vectorstore. Will take time, depending on the size of your document.
-You can ingest as many documents as you want by running `ingest`, and all will be accumulated in the local embeddings database. 
-If you want to start from scratch, delete the `db` folder.
+It will create a `db` folder containing the local vectorstore. Will take time, depending on the size of your documents.
+You can ingest as many documents as you want, and all will be accumulated in the local embeddings database. 
+If you want to start from an empty database, delete the `db` folder.
 
 Note: during the ingest process no data leaves your local environment. You could ingest without an internet connection.
 
@@ -59,7 +68,7 @@ Type `exit` to finish the script.
 Selecting the right local models and the power of `LangChain` you can run the entire pipeline locally, without any data leaving your environment, and with reasonable performance.
 
 - `ingest.py` uses `LangChain` tools to parse the document and create embeddings locally using `LlamaCppEmbeddings`. It then stores the result in a local vector database using `Chroma` vector store. 
-- `privateGPT.py` uses a local LLM based on `GPT4All-J` to understand questions and create answers. The context for the answers is extracted from the local vector store using a similarity search to locate the right piece of context from the docs.
+- `privateGPT.py` uses a local LLM based on `GPT4All-J` or `LlamaCpp` to understand questions and create answers. The context for the answers is extracted from the local vector store using a similarity search to locate the right piece of context from the docs.
 - `GPT4All-J` wrapper was introduced in LangChain 0.0.162.
 
 # Disclaimer

example.env

@@ -0,0 +1,5 @@
+PERSIST_DIRECTORY=db
+LLAMA_EMBEDDINGS_MODEL=models/ggml-model-q4_0.bin
+MODEL_TYPE=GPT4All
+MODEL_PATH=models/ggml-gpt4all-j-v1.3-groovy.bin
+MODEL_N_CTX=1000

ingest.py

@@ -1,19 +1,29 @@
-from langchain.document_loaders import TextLoader
+import os
+from langchain.document_loaders import TextLoader, PDFMinerLoader, CSVLoader
 from langchain.text_splitter import RecursiveCharacterTextSplitter
 from langchain.vectorstores import Chroma
 from langchain.embeddings import LlamaCppEmbeddings
-from sys import argv
 from constants import PERSIST_DIRECTORY
 from constants import CHROMA_SETTINGS
 
 def main():
+    llama_embeddings_model = os.environ.get('LLAMA_EMBEDDINGS_MODEL')
+    persist_directory = os.environ.get('PERSIST_DIRECTORY')
+    model_n_ctx = os.environ.get('MODEL_N_CTX')
     # Load document and split in chunks
-    loader = TextLoader(argv[1], encoding="utf8")
+    for root, dirs, files in os.walk("source_documents"):
+        for file in files:
+            if file.endswith(".txt"):
+                loader = TextLoader(os.path.join(root, file), encoding="utf8")
+            elif file.endswith(".pdf"):
+                loader = PDFMinerLoader(os.path.join(root, file))
+            elif file.endswith(".csv"):
+                loader = CSVLoader(os.path.join(root, file))
     documents = loader.load()
     text_splitter = RecursiveCharacterTextSplitter(chunk_size=500, chunk_overlap=50)
     texts = text_splitter.split_documents(documents)
     # Create embeddings
-    llama = LlamaCppEmbeddings(model_path="./models/ggml-model-q4_0.bin")
+    llama = LlamaCppEmbeddings(model_path=llama_embeddings_model, n_ctx=model_n_ctx)
     # Create and store locally vectorstore
     db = Chroma.from_documents(texts, llama, persist_directory=PERSIST_DIRECTORY, client_settings=CHROMA_SETTINGS)
     db.persist()

privateGPT.py

@@ -2,18 +2,32 @@ from langchain.chains import RetrievalQA
 from langchain.embeddings import LlamaCppEmbeddings
 from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
 from langchain.vectorstores import Chroma
-from langchain.llms import GPT4All
-from constants import PERSIST_DIRECTORY
+from langchain.llms import GPT4All, LlamaCpp
+import os
+
+llama_embeddings_model = os.environ.get("LLAMA_EMBEDDINGS_MODEL")
+persist_directory = os.environ.get('PERSIST_DIRECTORY')
+
+model_type = os.environ.get('MODEL_TYPE')
+model_path = os.environ.get('MODEL_PATH')
+model_n_ctx = os.environ.get('MODEL_N_CTX')
+
 from constants import CHROMA_SETTINGS
 
-def main():        
-    # Load stored vectorstore
-    llama = LlamaCppEmbeddings(model_path="./models/ggml-model-q4_0.bin")
-    db = Chroma(persist_directory=PERSIST_DIRECTORY, embedding_function=llama, client_settings=CHROMA_SETTINGS)
+def main():
+    llama = LlamaCppEmbeddings(model_path=llama_embeddings_model, n_ctx=model_n_ctx)
+    db = Chroma(persist_directory=persist_directory, embedding_function=llama, client_settings=CHROMA_SETTINGS)
     retriever = db.as_retriever()
     # Prepare the LLM
     callbacks = [StreamingStdOutCallbackHandler()]
-    llm = GPT4All(model='./models/ggml-gpt4all-j-v1.3-groovy.bin', backend='gptj', callbacks=callbacks, verbose=False)
+    match model_type:
+        case "LlamaCpp":
+            llm = LlamaCpp(model_path=model_path, n_ctx=model_n_ctx, callbacks=callbacks, verbose=False)
+        case "GPT4All":
+            llm = GPT4All(model=model_path, n_ctx=model_n_ctx, backend='gptj', callbacks=callbacks, verbose=False)
+        case _default:
+            print(f"Model {model_type} not supported!")
+            exit;
     qa = RetrievalQA.from_chain_type(llm=llm, chain_type="stuff", retriever=retriever, return_source_documents=True)
     # Interactive questions and answers
     while True: