Bugfixes and features

bootstraprag/templates/evaluations/ragas_evals/.env

@@ -1,12 +1,12 @@
 input_dir='data'
 collection_name='topic_node_parser'
-qdrant_url='http://localhost:6333'
+qdrant_url='http://localhost:6333/'
 qdrant_api_key='th3s3cr3tk3y'
 llm_url='http://localhost:11434'
 llm_model='llama3.2:latest'
 embed_model_name='nomic-embed-text:latest'
 sparse_model='Qdrant/bm42-all-minilm-l6-v2-attentions'
-OPENAI_API_KEY=<YOUR_API_KEY>
+OPENAI_API_KEY='sk-proj-'
 retrieve_top_k=5
 enable_postprocessing_method=false
 postprocessing_method='llm_reranker'

bootstraprag/templates/evaluations/ragas_evals/llamaindex/llamaindex_ragas.py

@@ -1,6 +1,6 @@
 import os
 import logging
-from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings
+from llama_index.core import VectorStoreIndex, SimpleDirectoryReader, Settings, StorageContext
 from llama_index.vector_stores.qdrant import QdrantVectorStore
 from llama_index.core.postprocessor import LLMRerank, SentenceTransformerRerank, LongContextReorder
 from llama_index.llms.openai import OpenAI
@@ -30,12 +30,16 @@ def __init__(self, input_dir="../data", model_name=os.environ.get("llm_model"),
         logger.info(
             f"Initializing LlamaIndexEvaluator with input_dir={input_dir}, model_name={model_name}, base_url={base_url}")
 
+        Settings.llm = Ollama(model=os.environ.get('llm_model'), base_url=base_url)
+        Settings.embed_model = OllamaEmbedding(model_name=os.environ.get('embed_model_name'), base_url=base_url)
+
         self.input_dir = input_dir
         self.model_name = model_name
         self.base_url = base_url
         self.documents = self.load_documents()
+        self.index_data(documents=self.documents)
         self.top_k = int(os.environ.get('retrieve_top_k'))
-        Settings.llm = Ollama(model=self.model_name, base_url=self.base_url)
+
         logger.info(f"Loading dataset from {os.environ.get('gold_dataset_file')}")
         self.dataset = pd.read_json(os.path.join(input_dir, os.environ.get('gold_dataset_file')))
         self.query_engine = self.build_query_engine(postprocessing_method=os.environ.get('postprocessing_method'))
@@ -48,6 +52,17 @@ def load_documents(self):
         logger.info(f"Loaded {len(documents)} documents")
         return documents
 
+    def index_data(self, documents):
+        logger.info("Indexing data in Qdrant")
+        # Initialize Qdrant client
+        qdrant_cli = qdrant_client.QdrantClient(url=os.environ.get("qdrant_url"),
+                                                api_key=os.environ.get("qdrant_api_key"))
+        qdrant_vector_store = QdrantVectorStore(collection_name=os.environ.get("collection_name"), client=qdrant_cli)
+        storage_context = StorageContext.from_defaults(vector_store=qdrant_vector_store)
+        if not qdrant_cli.collection_exists(collection_name=os.environ.get("collection_name")):
+            VectorStoreIndex.from_documents(storage_context=storage_context, documents=documents, show_progress=True)
+        logger.info("Indexing data in Qdrant finished")
+
     def build_query_engine(self, postprocessing_method: str):
         logger.info("Initializing Qdrant client")
         # Initialize Qdrant client
@@ -71,16 +86,18 @@ def build_query_engine(self, postprocessing_method: str):
         if os.environ.get('enable_postprocessing_method') == 'true' and postprocessing_method == 'llm_reranker':
             reranker = LLMRerank(llm=Settings.llm, choice_batch_size=self.top_k)
             query_engine = vector_index.as_query_engine(top_k=self.top_k, node_postprocessors=[reranker])
-        elif os.environ.get('enable_postprocessing_method') == 'true' and postprocessing_method == 'sentence_transformer_rerank':
+        elif os.environ.get(
+                'enable_postprocessing_method') == 'true' and postprocessing_method == 'sentence_transformer_rerank':
             reranker = SentenceTransformerRerank(
                 model="cross-encoder/ms-marco-MiniLM-L-2-v2", top_n=self.top_k
             )
             query_engine = vector_index.as_query_engine(top_k=self.top_k, node_postprocessors=[reranker])
-        elif os.environ.get('enable_postprocessing_method') == 'true' and postprocessing_method == 'long_context_reorder':
+        elif os.environ.get(
+                'enable_postprocessing_method') == 'true' and postprocessing_method == 'long_context_reorder':
             reorder = LongContextReorder()
             query_engine = vector_index.as_query_engine(top_k=self.top_k, node_postprocessors=[reorder])
-
-        query_engine = vector_index.as_query_engine(similarity_top_k=self.top_k, llm=Settings.llm)
+        else:
+            query_engine = vector_index.as_query_engine(similarity_top_k=self.top_k, llm=Settings.llm)
         logger.info(f"Query engine built successfully: {query_engine}")
         return query_engine
 

bootstraprag/templates/llamaindex/rag_with_controllable_agents_with_observability/docker-compose-langfuse.yml

@@ -2,11 +2,10 @@ version: '3.8'
 
 services:
   postgres:
-    image: postgres:13
+    image: postgres:latest
     container_name: postgres
     environment:
-      POSTGRES_USER: admin
-      POSTGRES_PASSWORD: admin@123
+      POSTGRES_PASSWORD: postgres
       POSTGRES_DB: langfuse
     volumes:
       - postgres_data:/var/lib/postgresql/data
@@ -17,7 +16,7 @@ services:
     image: langfuse/langfuse:latest
     container_name: langfuse
     environment:
-      DATABASE_URL: postgresql://hello:mysecretpassword@postgres:5432/langfuse
+      DATABASE_URL: postgresql://postgres:postgres@postgres:5432/langfuse
       NEXTAUTH_URL: http://localhost:3000
       NEXTAUTH_SECRET: mysecret
       SALT: mysalt

bootstraprag/templates/llamaindex/rag_with_hyde/.env

@@ -1,6 +1,6 @@
 DB_URL='http://localhost:6333'
 DB_API_KEY='th3s3cr3tk3y'
-COLLECTION_NAME='YOUR_COLLECTION'
+COLLECTION_NAME='HYDE_COLLECTION'
 
 OPENAI_API_KEY=''
 OPENAI_EMBED_MODEL=''

bootstraprag/templates/llamaindex/rag_with_hyde/base_rag.py

@@ -65,7 +65,7 @@ def __init__(self, data_path: str, chunk_size: int = 512, chunk_overlap: int = 2
         Settings.llm = llm
         Settings.transformations = [self.text_parser]
 
-        self.rag_evaluator = RAGEvaluator
+        self.rag_evaluator = RAGEvaluator()
 
         self.text_chunks = []
         self.doc_ids = []

bootstraprag/templates/llamaindex/rag_with_hyde_with_observability/docker-compose-langfuse.yml

@@ -2,11 +2,10 @@ version: '3.8'
 
 services:
   postgres:
-    image: postgres:13
+    image: postgres:latest
     container_name: postgres
     environment:
-      POSTGRES_USER: admin
-      POSTGRES_PASSWORD: admin@123
+      POSTGRES_PASSWORD: postgres
       POSTGRES_DB: langfuse
     volumes:
       - postgres_data:/var/lib/postgresql/data
@@ -17,7 +16,7 @@ services:
     image: langfuse/langfuse:latest
     container_name: langfuse
     environment:
-      DATABASE_URL: postgresql://hello:mysecretpassword@postgres:5432/langfuse
+      DATABASE_URL: postgresql://postgres:postgres@postgres:5432/langfuse
       NEXTAUTH_URL: http://localhost:3000
       NEXTAUTH_SECRET: mysecret
       SALT: mysalt

bootstraprag/templates/llamaindex/rag_with_mlflow/readme.md

@@ -0,0 +1 @@
+## In progress

bootstraprag/templates/llamaindex/rag_with_react_with_observability/docker-compose-langfuse.yml

@@ -2,11 +2,10 @@ version: '3.8'
 
 services:
   postgres:
-    image: postgres:13
+    image: postgres:latest
     container_name: postgres
     environment:
-      POSTGRES_USER: admin
-      POSTGRES_PASSWORD: admin@123
+      POSTGRES_PASSWORD: postgres
       POSTGRES_DB: langfuse
     volumes:
       - postgres_data:/var/lib/postgresql/data
@@ -17,7 +16,7 @@ services:
     image: langfuse/langfuse:latest
     container_name: langfuse
     environment:
-      DATABASE_URL: postgresql://hello:mysecretpassword@postgres:5432/langfuse
+      DATABASE_URL: postgresql://postgres:postgres@postgres:5432/langfuse
       NEXTAUTH_URL: http://localhost:3000
       NEXTAUTH_SECRET: mysecret
       SALT: mysalt

bootstraprag/templates/llamaindex/rag_with_self_correction_with_observability/docker-compose-langfuse.yml

@@ -2,11 +2,10 @@ version: '3.8'
 
 services:
   postgres:
-    image: postgres:13
+    image: postgres:latest
     container_name: postgres
     environment:
-      POSTGRES_USER: admin
-      POSTGRES_PASSWORD: admin@123
+      POSTGRES_PASSWORD: postgres
       POSTGRES_DB: langfuse
     volumes:
       - postgres_data:/var/lib/postgresql/data
@@ -17,7 +16,7 @@ services:
     image: langfuse/langfuse:latest
     container_name: langfuse
     environment:
-      DATABASE_URL: postgresql://hello:mysecretpassword@postgres:5432/langfuse
+      DATABASE_URL: postgresql://postgres:postgres@postgres:5432/langfuse
       NEXTAUTH_URL: http://localhost:3000
       NEXTAUTH_SECRET: mysecret
       SALT: mysalt

bootstraprag/templates/llamaindex/rag_with_vision/readme.md

@@ -0,0 +1 @@
+## In progress

bootstraprag/templates/llamaindex/simple_rag_with_observability/docker-compose-langfuse.yml

@@ -0,0 +1,36 @@
+version: '3.8'
+
+services:
+  postgres:
+    image: postgres:latest
+    container_name: postgres
+    environment:
+      POSTGRES_PASSWORD: postgres
+      POSTGRES_DB: langfuse
+    volumes:
+      - postgres_data:/var/lib/postgresql/data
+    networks:
+      - langfuse-network
+
+  langfuse:
+    image: langfuse/langfuse:latest
+    container_name: langfuse
+    environment:
+      DATABASE_URL: postgresql://postgres:postgres@postgres:5432/langfuse
+      NEXTAUTH_URL: http://localhost:3000
+      NEXTAUTH_SECRET: mysecret
+      SALT: mysalt
+      ENCRYPTION_KEY: 98637d42c277ef10b8a324e25d492daa8eee8f769574124ba25132f71481f183
+    ports:
+      - "3000:3000"
+    depends_on:
+      - postgres
+    networks:
+      - langfuse-network
+
+volumes:
+  postgres_data:
+
+networks:
+  langfuse-network:
+    driver: bridge

bootstraprag/templates/llamaindex/simple_rag_with_observability/requirements.txt

@@ -1,13 +1,13 @@
 python-dotenv==1.0.1
-llama-index==0.11.7
-llama-index-llms-openai==0.2.3
-llama-index-llms-ollama==0.3.1
-llama-index-embeddings-openai==0.2.4
-llama-index-embeddings-ollama==0.3.0
-llama-index-vector-stores-qdrant==0.3.0
-qdrant-client==1.11.1
-arize-phoenix==4.33.1
-llama-index-callbacks-arize-phoenix==0.2.1
-pydantic==2.9.0
-litserve==0.2.2
-deepeval==1.3.2
+llama-index==0.11.19
+llama-index-llms-openai==0.2.16
+llama-index-llms-ollama==0.3.4
+llama-index-embeddings-openai==0.2.5
+llama-index-embeddings-ollama==0.3.1
+llama-index-vector-stores-qdrant==0.3.2
+qdrant-client==1.12.0
+pydantic==2.9.2
+litserve==0.2.3
+deepeval==1.4.4
+# observability
+langfuse==2.52.2

bootstraprag/templates/llamaindex/simple_rag_with_observability/simple_rag.py

@@ -9,23 +9,22 @@
 from llama_index.core.agent import ReActAgent
 from llama_index.llms.ollama import Ollama
 from llama_index.core.base.response.schema import Response, StreamingResponse, AsyncStreamingResponse, PydanticResponse
+from langfuse.llama_index import LlamaIndexInstrumentor
 from rag_evaluator import RAGEvaluator
 from dotenv import load_dotenv, find_dotenv
 from typing import Union
 import qdrant_client
 import logging
 import os
-import llama_index
-import phoenix as px
 
 _ = load_dotenv(find_dotenv())
 
 logging.basicConfig(level=int(os.environ['INFO']))
 logger = logging.getLogger(__name__)
 
 # instrumenting observability
-session = px.launch_app()
-llama_index.core.set_global_handler("arize_phoenix")
+instrumentor = LlamaIndexInstrumentor()
+instrumentor.start()
 
 
 class SimpleRAG:

setup.py

@@ -6,7 +6,7 @@
 
 setup(
     name='bootstrap-rag',
-    version='0.0.11',
+    version='0.0.12',
     long_description=long_description,
     long_description_content_type="text/markdown",
     packages=find_packages(),