wizit-context-ingestor 0.2.3b0__tar.gz → 0.2.4b0__tar.gz

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: wizit-context-ingestor
-Version: 0.2.3b0
+Version: 0.2.4b0
 Summary: Contextual Rag with Cloud Solutions
 Requires-Dist: anthropic[vertex]>=0.66.0
 Requires-Dist: boto3>=1.40.23
@@ -79,6 +79,14 @@ from main import context_chunks_in_document
 # Get semantic chunks from a document
 context_chunks_in_document("your-document.pdf")
 ```
+## Running Memory Profiler
+
+To run the memory profiler, use the following command:
+
+```bash
+python -m memray run test_redis.py
+```
+
 
 ## Project Structure
 

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/README.md

@@ -64,6 +64,14 @@ from main import context_chunks_in_document
 # Get semantic chunks from a document
 context_chunks_in_document("your-document.pdf")
 ```
+## Running Memory Profiler
+
+To run the memory profiler, use the following command:
+
+```bash
+python -m memray run test_redis.py
+```
+
 
 ## Project Structure
 

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "wizit_context_ingestor"
-version = "0.2.3-beta"
+version = "0.2.4-beta"
 description = "Contextual Rag with Cloud Solutions"
 readme = "README.md"
 requires-python = ">=3.12"
@@ -17,6 +17,8 @@ dependencies = [
 
 [dependency-groups]
 dev = [
+    "memray>=1.18.0",
+    "pyinstrument>=5.1.1",
     "python-dotenv>=1.1.1",
 ]
 

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/aws_model.py

@@ -12,7 +12,7 @@ class AWSModels(AiApplicationService):
     A wrapper class for Google Cloud Vertex AI models that handles credentials and
     provides methods to load embeddings and chat models.
     """
-
+    __slots__ = ('llm_model_id')
     def __init__(
         self,
         llm_model_id: str = "us.anthropic.claude-3-5-sonnet-20241022-v2:0"

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/persistence/s3_storage.py

@@ -1,6 +1,6 @@
 from ...application.interfaces import PersistenceService
 from ...domain.models import ParsedDoc
-import boto3
+from boto3 import client as boto3_client
 import logging
 import os
 from botocore.exceptions import ClientError
@@ -10,9 +10,9 @@ logger = logging.getLogger(__name__)
 
 class S3StorageService(PersistenceService):
     """Persistence service for S3 storage."""
-
+    __slots__ = ('origin_bucket_name', 'target_bucket_name', 'region_name')
     def __init__(self, origin_bucket_name: str, target_bucket_name: str, region_name: str = 'us-east-1'):
-        self.s3 = boto3.client('s3', region_name=region_name)
+        self.s3 = boto3_client('s3', region_name=region_name)
         self.origin_bucket_name = origin_bucket_name
         self.target_bucket_name = target_bucket_name
 

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/rag/pg_embeddings.py

@@ -38,7 +38,7 @@ class PgEmbeddingsManager(EmbeddingsManager):
       ... )
       >>> documents = [Document(page_content="Sample text", metadata={"source": "example"})]
     """
-
+    __slots__ = ("embeddings_model", "pg_connection")
     def __init__(self, embeddings_model, pg_connection: str):
         """
           Initialize the PgEmbeddingsManager.

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/rag/redis_embeddings.py

@@ -11,6 +11,7 @@ logger = logging.getLogger(__name__)
 
 class RedisEmbeddingsManager(EmbeddingsManager):
 
+    __slots__ = ("embeddings_model", "redis_conn_string", "metadata_tags")
     def __init__(self, embeddings_model, redis_conn_string: str, metadata_tags: dict):
         """
         Initialize the RedisEmbeddingsManager.

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/rag/semantic_chunks.py

@@ -16,7 +16,7 @@ class SemanticChunks(RagChunker):
     Class for semantically chunking documents into smaller pieces based on semantic similarity.
     Uses LangChain's SemanticChunker to create semantically coherent document chunks.
     """
-
+    __slots__ = ("embeddings_model",)
     def __init__(self, embeddings_model: Any):
         """
         Initialize a document chunker with an embeddings model.

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/secrets/aws_secrets_manager.py

@@ -1,4 +1,4 @@
-import boto3
+from boto3 import client as boto3_client
 import logging
 
 logger = logging.getLogger(__name__)
@@ -6,7 +6,7 @@ logger = logging.getLogger(__name__)
 class AwsSecretsManager:
 
     def __init__(self):
-        self.client = boto3.client('secretsmanager')
+        self.client = boto3_client('secretsmanager')
 
 
     def get_secret(self, secret_name):
@@ -30,4 +30,4 @@ class AwsSecretsManager:
             return msg
         except Exception as e:
             logger.error(f"An unknown error occurred: {str(e)}.")
-            raise
+            raise

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/infra/vertex_model.py

@@ -1,4 +1,4 @@
-import vertexai
+from vertexai import init as vertexai_init
 from google.oauth2 import service_account
 from langchain_google_vertexai import VertexAIEmbeddings, ChatVertexAI
 from langchain_google_vertexai.model_garden import ChatAnthropicVertex
@@ -15,7 +15,7 @@ class VertexModels(AiApplicationService):
     A wrapper class for Google Cloud Vertex AI models that handles credentials and
     provides methods to load embeddings and chat models.
     """
-
+    __slots__ = ('project_id', 'location', 'json_service_account', 'scopes', 'llm_model_id')
     def __init__(
             self,
             project_id: str,
@@ -42,7 +42,7 @@ class VertexModels(AiApplicationService):
             self.llm_model_id = llm_model_id
             self.project_id = project_id
             self.location = location
-            vertexai.init(
+            vertexai_init(
                 project=project_id,
                 location=location,
                 credentials=self.credentials
@@ -54,7 +54,7 @@ class VertexModels(AiApplicationService):
 
     def load_embeddings_model(
         self,
-        embeddings_model_id: str = "text-embedding-005") -> VertexAIEmbeddings:  # noqa: E125
+        embeddings_model_id: str = "text-multilingual-embedding-002") -> VertexAIEmbeddings:  # noqa: E125
         """
         Load and return a Vertex AI embeddings model.
         default embeddings length is 768 https://cloud.google.com/vertex-ai/generative-ai/docs/embeddings/get-text-embeddings

{wizit_context_ingestor-0.2.3b0 → wizit_context_ingestor-0.2.4b0}/src/wizit_context_ingestor/main.py

@@ -78,6 +78,7 @@ class DeelabRedisChunksManager:
             gcp_secret_name: str,
             redis_connection_string: str,
             llm_model_id: str = "claude-3-5-haiku@20241022",
+            embeddings_model_id: str = "text-multilingual-embedding-002",
             target_language: str = "es"
     ):
         self.gcp_project_id = gcp_project_id
@@ -89,7 +90,7 @@ class DeelabRedisChunksManager:
         self.gcp_sa_dict = self._get_gcp_sa_dict(gcp_secret_name)
         self.redis_connection_string = redis_connection_string
         self.vertex_model = self._get_vertex_model()
-        self.embeddings_model = self.vertex_model.load_embeddings_model()
+        self.embeddings_model = self.vertex_model.load_embeddings_model(embeddings_model_id)
 
     def _get_gcp_sa_dict(self, gcp_secret_name: str):
         vertex_gcp_sa = self.aws_secrets_manager.get_secret(gcp_secret_name)