Updated libraries to the latest

.vscode/settings.json

@@ -1,10 +1,11 @@
 {
-    "[python]": {
-      "editor.formatOnSave": true,
-      "editor.defaultFormatter": "charliermarsh.ruff",
-      "editor.codeActionsOnSave": {
-        "source.fixAll": "never",
-        "source.organizeImports": "explicit"
-      }
+  "[python]": {
+    "editor.formatOnSave": true,
+    "editor.defaultFormatter": "charliermarsh.ruff",
+    "editor.codeActionsOnSave": {
+      "source.fixAll": "never",
+      "source.organizeImports": "explicit"
     }
-  }
+  },
+  "wolf.disableHotModeWarning": true
+}

notebooks/reranking-and-filtering.ipynb

@@ -275,7 +275,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "ragbase-YTjELISx-py3.12",
+   "display_name": "ragbase-BxuiWsRr-py3.12",
    "language": "python",
    "name": "python3"
   },
@@ -289,7 +289,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.4"
+   "version": "3.12.5"
   }
  },
  "nbformat": 4,

pyproject.toml

@@ -1,29 +1,29 @@
 [tool.poetry]
 name = "ragbase"
-version = "0.1.0"
+version = "0.1.1"
 description = ""
 authors = ["Venelin Valkov <[email protected]>"]
 license = "MIT"
 readme = "README.md"
 
 [tool.poetry.dependencies]
 python = ">=3.12,<3.13"
-langchain-community = "^0.2.6"
-flashrank = "^0.2.8"
-qdrant-client = "^1.10.1"
+langchain-community = "^0.3.29"
+flashrank = "^0.2.10"
+qdrant-client = "^1.15.1"
 pypdfium2 = "^4.30.0"
-fastembed = "^0.3.3"
-langchain-experimental = "^0.0.62"
-langchain-qdrant = "^0.1.1"
-streamlit = "^1.36.0"
-python-dotenv = "^1.0.1"
-langchain-groq = "^0.1.6"
+fastembed = "^0.7.3"
+langchain-experimental = "^0.3.4"
+langchain-qdrant = "^0.2.1"
+streamlit = "^1.49.1"
+python-dotenv = "^1.1.1"
+langchain-groq = "^0.3.8"
 
 
 [tool.poetry.group.dev.dependencies]
-ruff = "^0.5.1"
-jupyterlab = "^4.2.3"
-watchdog = "^4.0.1"
+ruff = "^0.13.1"
+jupyterlab = "^4.4.7"
+watchdog = "^6.0.0"
 
 [build-system]
 requires = ["poetry-core"]

ragbase/config.py

@@ -19,11 +19,11 @@ class Model:
         REMOTE_LLM = "llama-3.1-70b-versatile"
         TEMPERATURE = 0.0
         MAX_TOKENS = 8000
-        USE_LOCAL = False
+        USE_LOCAL = True
 
     class Retriever:
         USE_RERANKER = True
         USE_CHAIN_FILTER = False
 
-    DEBUG = False
-    CONVERSATION_MESSAGES_LIMIT = 6
+    DEBUG = True
+    CONVERSATION_MESSAGES_LIMIT = 100

ragbase/model.py

@@ -1,5 +1,6 @@
 from langchain_community.chat_models import ChatOllama
-from langchain_community.document_compressors.flashrank_rerank import FlashrankRerank
+from langchain_community.document_compressors.flashrank_rerank import \
+    FlashrankRerank
 from langchain_community.embeddings.fastembed import FastEmbedEmbeddings
 from langchain_core.language_models import BaseLanguageModel
 from langchain_groq import ChatGroq
@@ -24,7 +25,16 @@ def create_llm() -> BaseLanguageModel:
 
 
 def create_embeddings() -> FastEmbedEmbeddings:
-    return FastEmbedEmbeddings(model_name=Config.Model.EMBEDDINGS)
+    if Config.Model.USE_GPU:
+        if Config.Model.APPLE_SILICON:
+            print("Using Apple Silicon")
+            return FastEmbedEmbeddings(model_name=Config.Model.EMBEDDINGS, device="apple_silicon")
+        else:
+            print("Using GPU")
+            return FastEmbedEmbeddings(model_name=Config.Model.EMBEDDINGS, device="cuda")
+    else:
+        print("Using CPU")
+        return FastEmbedEmbeddings(model_name=Config.Model.EMBEDDINGS, device="cpu")
 
 
 def create_reranker() -> FlashrankRerank:

requirements.txt

@@ -1,11 +1,11 @@
-fastembed==0.3.3
-flashrank==0.2.8
-langchain-community==0.2.6
-langchain-experimental==0.0.62
-langchain-groq==0.1.6
-langchain-qdrant==0.1.1
-langchain==0.2.6
-python-dotenv==1.0.1
-qdrant-client==1.10.1
-streamlit==1.36.0
+fastembed==0.7.3
+flashrank==0.2.10
+langchain-community==0.3.29
+langchain-experimental==0.3.4
+langchain-groq==0.3.8
+langchain-qdrant==0.2.1
+langchain==0.3.27
+python-dotenv==1.1.1
+qdrant-client==1.15.1
+streamlit==1.49.1
 pypdfium2==4.30.0