Add embed_anything module and update pyproject.toml

.gitignore

@@ -17,7 +17,6 @@ build/
 develop-eggs/
 dist/
 eggs/
-lib/
 lib64/
 parts/
 sdist/

pyproject.toml

@@ -19,6 +19,7 @@ license = {file = "LICENSE"}
 
 [tool.maturin]
 features = ["pyo3/extension-module"]
-
+include = [{path = "lib/*.dll", format = ["sdist", "wheel"]}]
+python-source = "python"
 [project.urls]
 Homepage = "https://github.com/StarlightSearch/EmbedAnything/tree/main"

python/embed_anything/__init__.py

@@ -0,0 +1,5 @@
+from .embed_anything import *
+
+__doc__ = embed_anything.__doc__
+if hasattr(embed_anything, "__all__"):
+    __all__ = embed_anything.__all__

python/embed_anything/embed_anything.pyi

@@ -0,0 +1,47 @@
+from .embed_anything import *
+
+def embed_query(query: list[str], embeder: str) -> list[EmbedData]:
+    """
+    Embeds the given query and returns an EmbedData object.
+
+    ### Arguments:
+    - `query`: The query to embed.
+    - `embeder`: The name of the embedding model to use. Choose between "OpenAI" and "AllMiniLmL12V2"
+
+    ### Returns:
+    - An EmbedData object.
+    """
+def embed_file(file_path: str, embeder: str) -> list[EmbedData]:
+    """
+    Embeds the file at the given path and returns a list of EmbedData objects.
+
+    ### Arguments:
+    - `file_path`: The path to the file to embed.
+    - `embeder`: The name of the embedding model to use. Choose between "OpenAI" and "AllMiniLmL12V2"
+
+    ### Returns:
+    - A list of EmbedData objects.
+    """
+
+def embed_directory(file_path: str, embeder: str) -> list[EmbedData]:
+    """
+    Embeds all the files in the given directory and returns a list of EmbedData objects.
+
+    ### Arguments:
+    - `file_path`: The path to the directory containing the files to embed.
+    - `embeder`: The name of the embedding model to use. Choose between "OpenAI" and "AllMiniLmL12V2"
+
+    ### Returns:
+    - A list of EmbedData objects.
+    """
+
+class EmbedData:
+    """
+    Represents the data of an embedded file.
+
+    ### Attributes:
+    - `embedding`: The embedding of the file.
+    - `text`: The text for which the embedding is generated for.
+    """
+    embedding: list[float]
+    text: str

test.py → python/embed_anything/test.py

@@ -1,19 +1,10 @@
 import os
 import numpy as np
-# os.add_dll_directory(r'D:\test')
-from embed_anything import EmbedData
 import embed_anything
+from embed_anything import EmbedData
 from PIL import Image
 import time
 
-# data:list[EmbedData] = embed_anything.embed_file("test_files/TUe_SOP_AI_2.pdf", embeder= "Bert")
-
-# embeddings = np.array([data.embedding for data in data])
-
-
-# print(embeddings)
-# print("Time taken: ", end-start)
-
 start = time.time()
 data:list[EmbedData] = embed_anything.embed_directory("test_files", embeder= "Clip")