Python: Migrate to Ollama Python SDK (#7165)

### Motivation and Context

<!-- Thank you for your contribution to the semantic-kernel repo!
Please help reviewers and future users, providing the following
information:
  1. Why is this change required?
  2. What problem does it solve?
  3. What scenario does it contribute to?
  4. If it fixes an open issue, please link to the issue here.
-->

### Description

<!-- Describe your changes, the overall approach, the underlying design.
These notes will help understanding how your code works. Thanks! -->
Currently, the Ollama AI connector uses `aiohttp` sessions to interface
with the local Ollama LLM service. However, we have made the decision to
transition to the official Ollama Python SDK.

Related to: #6841,
#7134

Test coverage report:

![image](https://github.com/microsoft/semantic-kernel/assets/12570346/8d414f71-fc1b-4870-b038-0f7d2985b985)

MyPy warnings: all cleared
### Contribution Checklist

<!-- Before submitting this PR, please make sure: -->

- [x] The code builds clean without any errors or warnings
- [x] The PR follows the [SK Contribution
Guidelines](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md)
and the [pre-submission formatting
script](https://github.com/microsoft/semantic-kernel/blob/main/CONTRIBUTING.md#development-scripts)
raises no violations
- [x] All unit tests pass, and I have added new tests where possible
- [x] I didn't break anyone 😄

.github/workflows/python-integration-tests.yml

@@ -70,6 +70,17 @@ jobs:
           python -m pip install --upgrade pip setuptools wheel
           cd python
           poetry install --with tests
+      - name: Install Ollama
+        if: matrix.os == 'ubuntu-latest'
+        run: |
+          curl -fsSL https://ollama.com/install.sh | sh
+          ollama serve &
+          sleep 5
+      - name: Pull model in Ollama
+        if: matrix.os == 'ubuntu-latest'
+        run: |
+          ollama pull ${{ vars.OLLAMA_MODEL }}
+          ollama list
       - name: Run Integration Tests
         id: run_tests
         shell: bash
@@ -98,6 +109,7 @@ jobs:
           ACA_POOL_MANAGEMENT_ENDPOINT: ${{secrets.ACA_POOL_MANAGEMENT_ENDPOINT}}
           MISTRALAI_API_KEY: ${{secrets.MISTRALAI_API_KEY}}
           MISTRALAI_CHAT_MODEL_ID: ${{ vars.MISTRALAI_CHAT_MODEL_ID }}
+          OLLAMA_MODEL: "${{ matrix.os == 'ubuntu-latest' && vars.OLLAMA_MODEL || '' }}" # phi3
         run: |
           if ${{ matrix.os == 'ubuntu-latest' }}; then
             docker run -d --name redis-stack-server -p 6379:6379 redis/redis-stack-server:latest
@@ -139,6 +151,19 @@ jobs:
           python -m pip install --upgrade pip setuptools wheel
           cd python && poetry install --with tests
 
+      - name: Install Ollama
+        if: matrix.os == 'ubuntu-latest'
+        run: |
+          curl -fsSL https://ollama.com/install.sh | sh
+          ollama serve &
+          sleep 5
+
+      - name: Pull model in Ollama
+        if: matrix.os == 'ubuntu-latest'
+        run: |
+          ollama pull ${{ vars.OLLAMA_MODEL }}
+          ollama list
+
       - name: Run Integration Tests
         id: run_tests
         shell: bash
@@ -167,6 +192,7 @@ jobs:
           ACA_POOL_MANAGEMENT_ENDPOINT: ${{secrets.ACA_POOL_MANAGEMENT_ENDPOINT}}
           MISTRALAI_API_KEY: ${{secrets.MISTRALAI_API_KEY}}
           MISTRALAI_CHAT_MODEL_ID: ${{ vars.MISTRALAI_CHAT_MODEL_ID }}
+          OLLAMA_MODEL: "${{ matrix.os == 'ubuntu-latest' && vars.OLLAMA_MODEL || '' }}" # phi3
         run: |
           if ${{ matrix.os == 'ubuntu-latest' }}; then
             docker run -d --name redis-stack-server -p 6379:6379 redis/redis-stack-server:latest

python/mypy.ini

@@ -13,10 +13,6 @@ warn_untyped_fields = true
 [mypy-semantic_kernel]
 no_implicit_reexport = true
 
-[mypy-semantic_kernel.connectors.ai.ollama.*]
-ignore_errors = true
-# TODO (eavanvalkenburg): remove this: https://github.com/microsoft/semantic-kernel/issues/7134
-
 [mypy-semantic_kernel.memory.*]
 ignore_errors = true
 # TODO (eavanvalkenburg): remove this

python/pyproject.toml

@@ -54,6 +54,8 @@ pymilvus = { version = ">=2.3,<2.4.4", optional = true}
 milvus = { version = ">=2.3,<2.3.8", markers = 'sys_platform != "win32"', optional = true}
 # mistralai
 mistralai = { version = "^0.4.1", optional = true}
+# ollama
+ollama = { version = "^0.2.1", optional = true}
 # pinecone
 pinecone-client = { version = ">=3.0.0", optional = true}
 # postgres
@@ -89,6 +91,7 @@ azure-search-documents = {version = "11.6.0b4", allow-prereleases = true}
 azure-core = "^1.28.0"
 azure-cosmos = "^4.7.0"
 mistralai = "^0.4.1"
+ollama = "^0.2.1"
 transformers = { version = "^4.28.1", extras=["torch"]}
 sentence-transformers = "^2.2.2"
 
@@ -113,6 +116,8 @@ pymilvus = ">=2.3,<2.4.4"
 milvus = { version = ">=2.3,<2.3.8", markers = 'sys_platform != "win32"'}
 # mistralai
 mistralai = "^0.4.1"
+# ollama
+ollama = "^0.2.1"
 # mongodb
 motor = "^3.3.2"
 # pinecone
@@ -131,13 +136,14 @@ weaviate-client = ">=3.18,<5.0"
 
 # Extras are exposed to pip, this allows a user to easily add the right dependencies to their environment
 [tool.poetry.extras]
-all = ["transformers", "sentence-transformers", "qdrant-client", "chromadb", "pymilvus", "milvus","mistralai", "weaviate-client", "pinecone-client", "psycopg", "redis", "azure-ai-inference", "azure-search-documents", "azure-core", "azure-identity", "azure-cosmos", "usearch", "pyarrow", "ipykernel", "motor"]
+all = ["transformers", "sentence-transformers", "qdrant-client", "chromadb", "pymilvus", "milvus", "mistralai", "ollama", "weaviate-client", "pinecone-client", "psycopg", "redis", "azure-ai-inference", "azure-search-documents", "azure-core", "azure-identity", "azure-cosmos", "usearch", "pyarrow", "ipykernel", "motor"]
 
 azure = ["azure-ai-inference", "azure-search-documents", "azure-core", "azure-identity", "azure-cosmos", "msgraph-sdk"]
 chromadb = ["chromadb"]
 hugging_face = ["transformers", "sentence-transformers"]
 milvus = ["pymilvus", "milvus"]
 mistralai = ["mistralai"]
+ollama = ["ollama"]
 mongo = ["motor"]
 notebooks = ["ipykernel"]
 pinecone = ["pinecone-client"]

python/semantic_kernel/connectors/ai/ollama/ollama_prompt_execution_settings.py

@@ -2,25 +2,28 @@
 
 from typing import Any, Literal
 
-from pydantic import Field
-
 from semantic_kernel.connectors.ai.prompt_execution_settings import PromptExecutionSettings
 
 
 class OllamaPromptExecutionSettings(PromptExecutionSettings):
-    ai_model_id: str = Field("", serialization_alias="model")
+    """Settings for Ollama prompt execution."""
+
     format: Literal["json"] | None = None
     options: dict[str, Any] | None = None
-    stream: bool = False
 
 
 class OllamaTextPromptExecutionSettings(OllamaPromptExecutionSettings):
-    prompt: str | None = None
-    context: str | None = None
+    """Settings for Ollama text prompt execution."""
+
     system: str | None = None
     template: str | None = None
-    raw: bool = False
+    context: str | None = None
+    raw: bool | None = None
 
 
 class OllamaChatPromptExecutionSettings(OllamaPromptExecutionSettings):
-    messages: list[dict[str, str]] | None = None
+    """Settings for Ollama chat prompt execution."""
+
+
+class OllamaEmbeddingPromptExecutionSettings(OllamaPromptExecutionSettings):
+    """Settings for Ollama embedding prompt execution."""

python/semantic_kernel/connectors/ai/ollama/ollama_settings.py

@@ -0,0 +1,29 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+from typing import ClassVar
+
+from semantic_kernel.kernel_pydantic import KernelBaseSettings
+
+
+class OllamaSettings(KernelBaseSettings):
+    """Ollama settings.
+
+    The settings are first loaded from environment variables with
+    the prefix 'OLLAMA_'.
+    If the environment variables are not found, the settings can
+    be loaded from a .env file with the encoding 'utf-8'.
+    If the settings are not found in the .env file, the settings
+    are ignored; however, validation will fail alerting that the
+    settings are missing.
+
+    Required settings for prefix 'OLLAMA' are:
+    - model: str - Model name. (Env var OLLAMA_MODEL)
+
+    Optional settings for prefix 'OLLAMA' are:
+    - host: HttpsUrl - The endpoint of the Ollama service. (Env var OLLAMA_HOST)
+    """
+
+    env_prefix: ClassVar[str] = "OLLAMA_"
+
+    model: str
+    host: str | None = None

python/semantic_kernel/connectors/ai/ollama/services/ollama_base.py

@@ -0,0 +1,17 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+from abc import ABC
+
+from ollama import AsyncClient
+
+from semantic_kernel.kernel_pydantic import KernelBaseModel
+
+
+class OllamaBase(KernelBaseModel, ABC):
+    """Ollama service base.
+
+    Args:
+        client [AsyncClient]: An Ollama client to use for the service.
+    """
+
+    client: AsyncClient