ddi-fw 0.0.261__tar.gz → 0.0.263__tar.gz

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ddi_fw
-Version: 0.0.261
+Version: 0.0.263
 Summary: Do not use :)
 Author-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>
 Maintainer-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/pyproject.toml

@@ -6,7 +6,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "ddi_fw"
-version = "0.0.261"
+version = "0.0.263"
 description = "Do not use :)"
 readme = "README.md"
 authors = [

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/src/ddi_fw/langchain/__init__.py

@@ -1,5 +1,5 @@
 from ..langchain.embeddings import PoolingStrategy,SumPoolingStrategy,MeanPoolingStrategy,SentenceTransformerDecorator,PretrainedEmbeddings,SBertEmbeddings
-from .sentence_splitter import SentenceSplitter
+from .sentence_splitter import SentenceSplitter,PassthroughTextSplitter
 # from .storage import DataFrameToVectorDB, generate_embeddings
 from .faiss_storage import BaseVectorStoreManager, FaissVectorStoreManager,generate_embeddings
 from .chroma_storage import ChromaVectorStoreManager

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/src/ddi_fw/langchain/chroma_storage.py

@@ -153,8 +153,8 @@ class ChromaVectorStoreManager(BaseVectorStoreManager):
     #                 print(f"{page_content_column}, size:{len(split_docs_chunk)}")
 
     @staticmethod
-    def get_persist_dir(base_dir, suffix, config=None):
-        return f"{base_dir}"
+    def get_persist_dir(base_dir,id, suffix, config=None):
+        return f"{base_dir}/chroma_db/{id}"
 
     def generate_vector_store(self, docs: List[Document]):
         self.vector_store = Chroma(

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/src/ddi_fw/langchain/faiss_storage.py

@@ -39,7 +39,7 @@ class BaseVectorStoreManager(BaseModel):
         raise NotImplementedError("This method should be implemented by subclasses.")
     
     @staticmethod
-    def get_persist_dir(base_dir, suffix, config=None):
+    def get_persist_dir(base_dir, id ,suffix, config=None):
         raise NotImplementedError("Subclasses must implement get_persist_dir.")
      
 
@@ -66,8 +66,8 @@ class FaissVectorStoreManager(BaseVectorStoreManager):
     #     self.vector_store.add_documents(documents=docs, ids=uuids)
     
     @staticmethod
-    def get_persist_dir(base_dir, suffix, config=None):
-        return f"{base_dir}/{suffix}"
+    def get_persist_dir(base_dir,id, suffix, config=None):
+        return f"{base_dir}/faiss/{id}/{suffix}"
     
     def initialize_embedding_dict(self, **kwargs):
         """
@@ -414,8 +414,9 @@ def generate_embeddings(
         # Load embedding model
         try:
             model_kwargs = collection_config.get('model_kwargs')
+            kwargs = {"model_kwargs":model_kwargs}
             model = get_import(embedding_model_type)(
-                model_name=name, **model_kwargs)
+                model_name=name, **kwargs)
         except Exception as e:
             raise Exception(f"Unknown embedding model: {embedding_model_type}") from e
 
@@ -441,7 +442,7 @@ def generate_embeddings(
             print(f"{id}_{suffix}")
             # persist_dir = f'{persist_directory}/{id}/{suffix}'
             # persist_dir = f'{persist_directory}/{suffix}'
-            persist_dir = vector_store_manager_type.get_persist_dir(persist_directory, suffix, collection_config)
+            persist_dir = vector_store_manager_type.get_persist_dir(persist_directory , id, suffix, collection_config)
 
             # Prepare manager parameters
             manager_params = {

ddi_fw-0.0.263/src/ddi_fw/langchain/sentence_splitter.py

@@ -0,0 +1,17 @@
+from typing import List
+import nltk
+from nltk import sent_tokenize
+from langchain_text_splitters.base import TextSplitter
+
+nltk.download('punkt')
+
+''' A text splitter that splits text into sentences using NLTK's sentence tokenizer.'''
+class SentenceSplitter(TextSplitter):
+    def split_text(self, text: str) -> List[str]:
+        return sent_tokenize(text)
+
+
+''' A text splitter that does not split the text at all, returning the entire text as a single chunk.'''
+class PassthroughTextSplitter(TextSplitter):
+    def split_text(self, text: str) -> List[str]:
+        return [text]

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/src/ddi_fw/utils/kaggle.py

@@ -37,6 +37,7 @@ def create_kaggle_dataset(base_path: str, collections: list):
         
         # Ensure title is between 6 and 50 characters
         if not (6 <= len(title) <= 50):
+            raise ValueError(f"Title length for {title} must be between 6 and 50 characters.")
             continue  # Skip if title length is out of the expected range
 
         # Step 3: Define the metadata content

{ddi_fw-0.0.261 → ddi_fw-0.0.263}/src/ddi_fw.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ddi_fw
-Version: 0.0.261
+Version: 0.0.263
 Summary: Do not use :)
 Author-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>
 Maintainer-email: Kıvanç Bayraktar <bayraktarkivanc@gmail.com>

ddi_fw-0.0.261/src/ddi_fw/langchain/sentence_splitter.py

@@ -1,10 +0,0 @@
-from typing import List
-import nltk
-from nltk import sent_tokenize
-from langchain_text_splitters.base import TextSplitter
-
-nltk.download('punkt')
-
-class SentenceSplitter(TextSplitter):
-    def split_text(self, text: str) -> List[str]:
-        return sent_tokenize(text)