Merge branch 'main' into main

microsoft · Jun 11, 2024 · 6a57096 · 6a57096
2 parents 735cf34 + bf7e4d6
commit 6a57096
Show file tree

Hide file tree

Showing 100 changed files with 3,744 additions and 424 deletions.
diff --git a/OAI_CONFIG_LIST_sample b/OAI_CONFIG_LIST_sample
@@ -13,13 +13,13 @@
         "api_key": "<your Azure OpenAI API key here>",
         "base_url": "<your Azure OpenAI API base here>",
         "api_type": "azure",
-        "api_version": "2024-02-15-preview"
+        "api_version": "2024-02-01"
     },
     {
         "model": "<your Azure OpenAI deployment name>",
         "api_key": "<your Azure OpenAI API key here>",
         "base_url": "<your Azure OpenAI API base here>",
         "api_type": "azure",
-        "api_version": "2024-02-15-preview"
+        "api_version": "2024-02-01"
     }
 ]
diff --git a/autogen/agentchat/contrib/vectordb/pgvectordb.py b/autogen/agentchat/contrib/vectordb/pgvectordb.py
@@ -32,10 +32,11 @@ class Collection:
         client: The PGVector client.
         collection_name (str): The name of the collection. Default is "documents".
         embedding_function (Callable): The embedding function used to generate the vector representation.
+            Default is None. SentenceTransformer("all-MiniLM-L6-v2").encode will be used when None.
+            Models can be chosen from:
+            https://huggingface.co/models?library=sentence-transformers
         metadata (Optional[dict]): The metadata of the collection.
         get_or_create (Optional): The flag indicating whether to get or create the collection.
-        model_name: (Optional str) | Sentence embedding model to use. Models can be chosen from:
-            https://huggingface.co/models?library=sentence-transformers
     """
 
     def __init__(
@@ -45,7 +46,6 @@ def __init__(
         embedding_function: Callable = None,
         metadata=None,
         get_or_create=None,
-        model_name="all-MiniLM-L6-v2",
     ):
         """
         Initialize the Collection object.
@@ -56,30 +56,26 @@ def __init__(
             embedding_function: The embedding function used to generate the vector representation.
             metadata: The metadata of the collection.
             get_or_create: The flag indicating whether to get or create the collection.
-            model_name: | Sentence embedding model to use. Models can be chosen from:
-                https://huggingface.co/models?library=sentence-transformers
         Returns:
             None
         """
         self.client = client
-        self.embedding_function = embedding_function
-        self.model_name = model_name
         self.name = self.set_collection_name(collection_name)
         self.require_embeddings_or_documents = False
         self.ids = []
-        try:
-            self.embedding_function = (
-                SentenceTransformer(self.model_name) if embedding_function is None else embedding_function
-            )
-        except Exception as e:
-            logger.error(
-                f"Validate the model name entered: {self.model_name} "
-                f"from https://huggingface.co/models?library=sentence-transformers\nError: {e}"
-            )
-            raise e
+        if embedding_function:
+            self.embedding_function = embedding_function
+        else:
+            self.embedding_function = SentenceTransformer("all-MiniLM-L6-v2").encode
         self.metadata = metadata if metadata else {"hnsw:space": "ip", "hnsw:construction_ef": 32, "hnsw:M": 16}
         self.documents = ""
         self.get_or_create = get_or_create
+        # This will get the model dimension size by computing the embeddings dimensions
+        sentences = [
+            "The weather is lovely today in paradise.",
+        ]
+        embeddings = self.embedding_function(sentences)
+        self.dimension = len(embeddings[0])
 
     def set_collection_name(self, collection_name) -> str:
         name = re.sub("-", "_", collection_name)
@@ -115,14 +111,14 @@ def add(self, ids: List[ItemID], documents: List, embeddings: List = None, metad
         elif metadatas is not None:
             for doc_id, metadata, document in zip(ids, metadatas, documents):
                 metadata = re.sub("'", '"', str(metadata))
-                embedding = self.embedding_function.encode(document)
+                embedding = self.embedding_function(document)
                 sql_values.append((doc_id, metadata, embedding, document))
             sql_string = (
                 f"INSERT INTO {self.name} (id, metadatas, embedding, documents)\n" f"VALUES (%s, %s, %s, %s);\n"
             )
         else:
             for doc_id, document in zip(ids, documents):
-                embedding = self.embedding_function.encode(document)
+                embedding = self.embedding_function(document)
                 sql_values.append((doc_id, document, embedding))
             sql_string = f"INSERT INTO {self.name} (id, documents, embedding)\n" f"VALUES (%s, %s, %s);\n"
         logger.debug(f"Add SQL String:\n{sql_string}\n{sql_values}")
@@ -166,7 +162,7 @@ def upsert(self, ids: List[ItemID], documents: List, embeddings: List = None, me
         elif metadatas is not None:
             for doc_id, metadata, document in zip(ids, metadatas, documents):
                 metadata = re.sub("'", '"', str(metadata))
-                embedding = self.embedding_function.encode(document)
+                embedding = self.embedding_function(document)
                 sql_values.append((doc_id, metadata, embedding, document, metadata, document, embedding))
             sql_string = (
                 f"INSERT INTO {self.name} (id, metadatas, embedding, documents)\n"
@@ -176,7 +172,7 @@ def upsert(self, ids: List[ItemID], documents: List, embeddings: List = None, me
             )
         else:
             for doc_id, document in zip(ids, documents):
-                embedding = self.embedding_function.encode(document)
+                embedding = self.embedding_function(document)
                 sql_values.append((doc_id, document, embedding, document))
             sql_string = (
                 f"INSERT INTO {self.name} (id, documents, embedding)\n"
@@ -304,7 +300,7 @@ def get(
                 )
         except (psycopg.errors.UndefinedTable, psycopg.errors.UndefinedColumn) as e:
             logger.info(f"Error executing select on non-existent table: {self.name}. Creating it instead. Error: {e}")
-            self.create_collection(collection_name=self.name)
+            self.create_collection(collection_name=self.name, dimension=self.dimension)
             logger.info(f"Created table {self.name}")
 
         cursor.close()
@@ -419,7 +415,7 @@ def query(
         cursor = self.client.cursor()
         results = []
         for query_text in query_texts:
-            vector = self.embedding_function.encode(query_text, convert_to_tensor=False).tolist()
+            vector = self.embedding_function(query_text, convert_to_tensor=False).tolist()
             if distance_type.lower() == "cosine":
                 index_function = "<=>"
             elif distance_type.lower() == "euclidean":
@@ -526,22 +522,31 @@ def delete_collection(self, collection_name: Optional[str] = None) -> None:
         cursor.execute(f"DROP TABLE IF EXISTS {self.name}")
         cursor.close()
 
-    def create_collection(self, collection_name: Optional[str] = None) -> None:
+    def create_collection(
+        self, collection_name: Optional[str] = None, dimension: Optional[Union[str, int]] = None
+    ) -> None:
         """
         Create a new collection.
 
         Args:
             collection_name (Optional[str]): The name of the new collection.
+            dimension (Optional[Union[str, int]]): The dimension size of the sentence embedding model
 
         Returns:
             None
         """
         if collection_name:
             self.name = collection_name
+
+        if dimension:
+            self.dimension = dimension
+        elif self.dimension is None:
+            self.dimension = 384
+
         cursor = self.client.cursor()
         cursor.execute(
             f"CREATE TABLE {self.name} ("
-            f"documents text, id CHAR(8) PRIMARY KEY, metadatas JSONB, embedding vector(384));"
+            f"documents text, id CHAR(8) PRIMARY KEY, metadatas JSONB, embedding vector({self.dimension}));"
             f"CREATE INDEX "
             f'ON {self.name} USING hnsw (embedding vector_l2_ops) WITH (m = {self.metadata["hnsw:M"]}, '
             f'ef_construction = {self.metadata["hnsw:construction_ef"]});'
@@ -573,7 +578,6 @@ def __init__(
         connect_timeout: Optional[int] = 10,
         embedding_function: Callable = None,
         metadata: Optional[dict] = None,
-        model_name: Optional[str] = "all-MiniLM-L6-v2",
     ) -> None:
         """
         Initialize the vector database.
@@ -591,15 +595,14 @@ def __init__(
             username: str | The database username to use. Default is None.
             password: str | The database user password to use. Default is None.
             connect_timeout: int | The timeout to set for the connection. Default is 10.
-            embedding_function: Callable | The embedding function used to generate the vector representation
-                of the documents. Default is None.
+            embedding_function: Callable | The embedding function used to generate the vector representation.
+                Default is None. SentenceTransformer("all-MiniLM-L6-v2").encode will be used when None.
+                Models can be chosen from:
+                https://huggingface.co/models?library=sentence-transformers
             metadata: dict | The metadata of the vector database. Default is None. If None, it will use this
                 setting: {"hnsw:space": "ip", "hnsw:construction_ef": 30, "hnsw:M": 16}. Creates Index on table
                 using hnsw (embedding vector_l2_ops) WITH (m = hnsw:M) ef_construction = "hnsw:construction_ef".
                 For more info: https://github.com/pgvector/pgvector?tab=readme-ov-file#hnsw
-            model_name: str | Sentence embedding model to use. Models can be chosen from:
-                https://huggingface.co/models?library=sentence-transformers
-
         Returns:
             None
         """
@@ -613,17 +616,10 @@ def __init__(
             password=password,
             connect_timeout=connect_timeout,
         )
-        self.model_name = model_name
-        try:
-            self.embedding_function = (
-                SentenceTransformer(self.model_name) if embedding_function is None else embedding_function
-            )
-        except Exception as e:
-            logger.error(
-                f"Validate the model name entered: {self.model_name} "
-                f"from https://huggingface.co/models?library=sentence-transformers\nError: {e}"
-            )
-            raise e
+        if embedding_function:
+            self.embedding_function = embedding_function
+        else:
+            self.embedding_function = SentenceTransformer("all-MiniLM-L6-v2").encode
         self.metadata = metadata
         register_vector(self.client)
         self.active_collection = None
@@ -738,7 +734,6 @@ def create_collection(
                 embedding_function=self.embedding_function,
                 get_or_create=get_or_create,
                 metadata=self.metadata,
-                model_name=self.model_name,
             )
             collection.set_collection_name(collection_name=collection_name)
             collection.create_collection(collection_name=collection_name)
@@ -751,7 +746,6 @@ def create_collection(
                 embedding_function=self.embedding_function,
                 get_or_create=get_or_create,
                 metadata=self.metadata,
-                model_name=self.model_name,
             )
             collection.set_collection_name(collection_name=collection_name)
             collection.create_collection(collection_name=collection_name)
@@ -765,7 +759,6 @@ def create_collection(
                 embedding_function=self.embedding_function,
                 get_or_create=get_or_create,
                 metadata=self.metadata,
-                model_name=self.model_name,
             )
             collection.set_collection_name(collection_name=collection_name)
             collection.create_collection(collection_name=collection_name)
@@ -797,7 +790,6 @@ def get_collection(self, collection_name: str = None) -> Collection:
                     client=self.client,
                     collection_name=collection_name,
                     embedding_function=self.embedding_function,
-                    model_name=self.model_name,
                 )
         return self.active_collection
 

diff --git a/autogen/oai/client.py b/autogen/oai/client.py
@@ -349,7 +349,7 @@ def __init__(self, *, config_list: Optional[List[Dict[str, Any]]] = None, **base
                 "api_key": os.environ.get("AZURE_OPENAI_API_KEY"),
                 "api_type": "azure",
                 "base_url": os.environ.get("AZURE_OPENAI_API_BASE"),
-                "api_version": "2024-02-15-preview",
+                "api_version": "2024-02-01",
             },
             {
                 "model": "gpt-3.5-turbo",
@@ -559,7 +559,7 @@ def yes_or_no_filter(context, response):
         ```
 
             - allow_format_str_template (bool | None): Whether to allow format string template in the config. Default to false.
-            - api_version (str | None): The api version. Default to None. E.g., "2024-02-15-preview".
+            - api_version (str | None): The api version. Default to None. E.g., "2024-02-01".
         Raises:
             - RuntimeError: If all declared custom model clients are not registered
             - APIError: If any model client create call raises an APIError

diff --git a/autogen/oai/completion.py b/autogen/oai/completion.py
@@ -741,7 +741,7 @@ def create(
                     "api_key": os.environ.get("AZURE_OPENAI_API_KEY"),
                     "api_type": "azure",
                     "base_url": os.environ.get("AZURE_OPENAI_API_BASE"),
-                    "api_version": "2024-02-15-preview",
+                    "api_version": "2024-02-01",
                 },
                 {
                     "model": "gpt-3.5-turbo",

diff --git a/autogen/oai/openai_utils.py b/autogen/oai/openai_utils.py
@@ -14,7 +14,7 @@
 from packaging.version import parse
 
 NON_CACHE_KEY = ["api_key", "base_url", "api_type", "api_version", "azure_ad_token", "azure_ad_token_provider"]
-DEFAULT_AZURE_API_VERSION = "2024-02-15-preview"
+DEFAULT_AZURE_API_VERSION = "2024-02-01"
 OAI_PRICE1K = {
     # https://openai.com/api/pricing/
     # gpt-4o
@@ -127,7 +127,7 @@ def get_config_list(
 
     # Optionally, define the API type and version if they are common for all keys
     api_type = 'azure'
-    api_version = '2024-02-15-preview'
+    api_version = '2024-02-01'
 
     # Call the get_config_list function to get a list of configuration dictionaries
     config_list = get_config_list(api_keys, base_urls, api_type, api_version)

diff --git a/autogen/version.py b/autogen/version.py
@@ -1 +1 @@
-__version__ = "0.2.28"
+__version__ = "0.2.29"
diff --git a/dotnet/AutoGen.sln b/dotnet/AutoGen.sln
@@ -53,6 +53,11 @@ Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "AutoGen.Anthropic.Tests", "
 EndProject
 Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "AutoGen.Anthropic.Samples", "sample\AutoGen.Anthropic.Samples\AutoGen.Anthropic.Samples.csproj", "{834B4E85-64E5-4382-8465-548F332E5298}"
 EndProject
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "AutoGen.Gemini", "src\AutoGen.Gemini\AutoGen.Gemini.csproj", "{EFE0DC86-80FC-4D52-95B7-07654BA1A769}"
+EndProject
+Project("{9A19103F-16F7-4668-BE54-9A1E7A4F7556}") = "AutoGen.Gemini.Tests", "test\AutoGen.Gemini.Tests\AutoGen.Gemini.Tests.csproj", "{8EA16BAB-465A-4C07-ABC4-1070D40067E9}"
+EndProject
+Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "AutoGen.Gemini.Sample", "sample\AutoGen.Gemini.Sample\AutoGen.Gemini.Sample.csproj", "{19679B75-CE3A-4DF0-A3F0-CA369D2760A4}"
 Project("{FAE04EC0-301F-11D3-BF4B-00C04F79EFBC}") = "AutoGen.AotCompatibility.Tests", "test\AutoGen.AotCompatibility.Tests\AutoGen.AotCompatibility.Tests.csproj", "{6B82F26D-5040-4453-B21B-C8D1F913CE4C}"
 EndProject
 Global
@@ -149,6 +154,18 @@ Global
 		{834B4E85-64E5-4382-8465-548F332E5298}.Debug|Any CPU.Build.0 = Debug|Any CPU
 		{834B4E85-64E5-4382-8465-548F332E5298}.Release|Any CPU.ActiveCfg = Release|Any CPU
 		{834B4E85-64E5-4382-8465-548F332E5298}.Release|Any CPU.Build.0 = Release|Any CPU
+		{EFE0DC86-80FC-4D52-95B7-07654BA1A769}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{EFE0DC86-80FC-4D52-95B7-07654BA1A769}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{EFE0DC86-80FC-4D52-95B7-07654BA1A769}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{EFE0DC86-80FC-4D52-95B7-07654BA1A769}.Release|Any CPU.Build.0 = Release|Any CPU
+		{8EA16BAB-465A-4C07-ABC4-1070D40067E9}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{8EA16BAB-465A-4C07-ABC4-1070D40067E9}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{8EA16BAB-465A-4C07-ABC4-1070D40067E9}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{8EA16BAB-465A-4C07-ABC4-1070D40067E9}.Release|Any CPU.Build.0 = Release|Any CPU
+		{19679B75-CE3A-4DF0-A3F0-CA369D2760A4}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
+		{19679B75-CE3A-4DF0-A3F0-CA369D2760A4}.Debug|Any CPU.Build.0 = Debug|Any CPU
+		{19679B75-CE3A-4DF0-A3F0-CA369D2760A4}.Release|Any CPU.ActiveCfg = Release|Any CPU
+		{19679B75-CE3A-4DF0-A3F0-CA369D2760A4}.Release|Any CPU.Build.0 = Release|Any CPU
 		{6B82F26D-5040-4453-B21B-C8D1F913CE4C}.Debug|Any CPU.ActiveCfg = Debug|Any CPU
 		{6B82F26D-5040-4453-B21B-C8D1F913CE4C}.Debug|Any CPU.Build.0 = Debug|Any CPU
 		{6B82F26D-5040-4453-B21B-C8D1F913CE4C}.Release|Any CPU.ActiveCfg = Release|Any CPU
@@ -180,6 +197,9 @@ Global
 		{6A95E113-B824-4524-8F13-CD0C3E1C8804} = {18BF8DD7-0585-48BF-8F97-AD333080CE06}
 		{815E937E-86D6-4476-9EC6-B7FBCBBB5DB6} = {F823671B-3ECA-4AE6-86DA-25E920D3FE64}
 		{834B4E85-64E5-4382-8465-548F332E5298} = {FBFEAD1F-29EB-4D99-A672-0CD8473E10B9}
+		{EFE0DC86-80FC-4D52-95B7-07654BA1A769} = {18BF8DD7-0585-48BF-8F97-AD333080CE06}
+		{8EA16BAB-465A-4C07-ABC4-1070D40067E9} = {F823671B-3ECA-4AE6-86DA-25E920D3FE64}
+		{19679B75-CE3A-4DF0-A3F0-CA369D2760A4} = {FBFEAD1F-29EB-4D99-A672-0CD8473E10B9}
 		{6B82F26D-5040-4453-B21B-C8D1F913CE4C} = {F823671B-3ECA-4AE6-86DA-25E920D3FE64}
 	EndGlobalSection
 	GlobalSection(ExtensibilityGlobals) = postSolution

diff --git a/dotnet/Directory.Build.props b/dotnet/Directory.Build.props
@@ -13,12 +13,37 @@
     <CSNoWarn>CS1998;CS1591</CSNoWarn>
     <NoWarn>$(NoWarn);$(CSNoWarn);NU5104</NoWarn>
     <TreatWarningsAsErrors>true</TreatWarningsAsErrors>
+    <GenerateDocumentationFile>true</GenerateDocumentationFile>
     <IsPackable>false</IsPackable>
     <EnableNetAnalyzers>true</EnableNetAnalyzers>
     <EnforceCodeStyleInBuild>true</EnforceCodeStyleInBuild>
+    <IsTestProject>false</IsTestProject>
   </PropertyGroup>
 
   <PropertyGroup>
     <RepoRoot>$(MSBuildThisFileDirectory)</RepoRoot>
   </PropertyGroup>
+
+  <ItemGroup Condition="'$(IsTestProject)' == 'true'">
+    <PackageReference Include="ApprovalTests" Version="$(ApprovalTestVersion)" />
+    <PackageReference Include="FluentAssertions" Version="$(FluentAssertionVersion)" />
+    <PackageReference Include="Microsoft.NET.Test.Sdk" Version="$(MicrosoftNETTestSdkVersion)" />
+    <PackageReference Include="xunit" Version="$(XUnitVersion)" />
+    <PackageReference Include="xunit.runner.console" Version="$(XUnitVersion)" />
+    <PackageReference Include="xunit.runner.visualstudio" Version="$(XUnitVersion)" />
+  </ItemGroup>
+
+  <ItemGroup  Condition="'$(IsTestProject)' == 'true'">
+    <Content Include="$(RepoRoot)resource/**/*.*">
+      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
+      <Link>testData/%(RecursiveDir)%(Filename)%(Extension)</Link>
+    </Content>
+  </ItemGroup>
+
+  <ItemGroup  Condition="'$(IncludeResourceFolder)' == 'true'">
+    <Content Include="$(RepoRoot)resource/**/*.*">
+      <CopyToOutputDirectory>Always</CopyToOutputDirectory>
+      <Link>resource/%(RecursiveDir)%(Filename)%(Extension)</Link>
+    </Content>
+  </ItemGroup>
 </Project>
diff --git a/dotnet/eng/Version.props b/dotnet/eng/Version.props
@@ -12,6 +12,7 @@
         <MicrosoftNETTestSdkVersion>17.7.0</MicrosoftNETTestSdkVersion>
         <MicrosoftDotnetInteractive>1.0.0-beta.24229.4</MicrosoftDotnetInteractive>
         <MicrosoftSourceLinkGitHubVersion>8.0.0</MicrosoftSourceLinkGitHubVersion>
+        <GoogleCloudAPIPlatformVersion>3.0.0</GoogleCloudAPIPlatformVersion>
         <JsonSchemaVersion>4.3.0.2</JsonSchemaVersion>
     </PropertyGroup>
 </Project>
diff --git a/...toGen.Ollama.Sample/images/background.png → dotnet/resource/images/background.png b/...toGen.Ollama.Sample/images/background.png → dotnet/resource/images/background.png
diff --git a/dotnet/resource/images/square.png b/dotnet/resource/images/square.png