gmicloud 0.1.3__tar.gz → 0.1.5__tar.gz

{gmicloud-0.1.3 → gmicloud-0.1.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: gmicloud
-Version: 0.1.3
+Version: 0.1.5
 Summary: GMI Cloud Python SDK
 Author-email: GMI <gmi@gmitec.net>
 License: MIT
@@ -13,16 +13,11 @@ Description-Content-Type: text/markdown
 # GMICloud SDK (Beta)
 
 ## Overview
-
 Before you start: Our service and GPU resource is currenly invite-only so please contact our team (getstarted@gmicloud.ai) to get invited if you don't have one yet.
 
-The GMI Inference Engine SDK provides a Python interface for deploying and managing machine learning models in
-production environments. It allows users to create model artifacts, schedule tasks for serving models, and call
-inference APIs easily.
+The GMI Inference Engine SDK provides a Python interface for deploying and managing machine learning models in production environments. It allows users to create model artifacts, schedule tasks for serving models, and call inference APIs easily.
 
-This SDK streamlines the process of utilizing GMI Cloud capabilities such as deploying models with Kubernetes-based Ray
-services, managing resources automatically, and accessing model inference endpoints. With minimal setup, developers can
-focus on building ML solutions instead of infrastructure.
+This SDK streamlines the process of utilizing GMI Cloud capabilities such as deploying models with Kubernetes-based Ray services, managing resources automatically, and accessing model inference endpoints. With minimal setup, developers can focus on building ML solutions instead of infrastructure.
 
 ## Features
 
@@ -50,6 +45,7 @@ Set the following environment variables:
 export GMI_CLOUD_CLIENT_ID=<YOUR_CLIENT_ID>
 export GMI_CLOUD_EMAIL=<YOUR_EMAIL>
 export GMI_CLOUD_PASSWORD=<YOUR_PASSWORD>
+export GMI_CLOUD_API_KEY=<YOUR_API_KEY>
 ```
 
 ### Option 2: Passing Credentials as Parameters
@@ -64,7 +60,18 @@ client = Client(client_id="<YOUR_CLIENT_ID>", email="<YOUR_EMAIL>", password="<Y
 
 ## Quick Start
 
-### 1. Create a Task from an Artifact Template
+### 1. How to run the code in the example folder
+```bash
+cd path/to/gmicloud-sdk
+# Create a virtual environment
+python -m venv venv
+source venv/bin/activate
+
+pip install -r requirements.txt
+python -m examples.create_task_from_artifact_template.py
+```
+
+### 2. Create a Task from an Artifact Template
 
 This is the simplest example to deploy an existing artifact template:
 
@@ -93,24 +100,31 @@ response = call_chat_completion(client, task.task_id)
 print(response)
 ```
 
-### 2. Step-by-Step Example: Create Artifact, Task, and Query the Endpoint
+### 3. Step-by-Step Example: Create Artifact, Task, and Query the Endpoint
 
 #### (a) Create an Artifact from a Template
 
 First, you’ll retrieve all templates and create an artifact based on the desired template (e.g., "Llama3.1 8B"):
 
 ```python
-def create_artifact_from_template(client):
+from gmicloud import *
+
+
+def create_artifact_from_template(client: Client) -> str:
     artifact_manager = client.artifact_manager
 
-    # List all available templates
-    templates = artifact_manager.get_artifact_templates()
+    # Get all artifact templates
+    templates = artifact_manager.get_public_templates()
     for template in templates:
         if template.artifact_template_id == "qwen_2.5_14b_instruct_template_001":
-            return artifact_manager.create_artifact_from_template(
-                artifact_template_id=template.artifact_template_id
+            # Create an artifact from a template
+            artifact_id = artifact_manager.create_artifact_from_template(
+                artifact_template_id=template.artifact_template_id,
             )
-    return None
+
+            return artifact_id
+
+    return ""
 ```
 
 #### (b) Create a Task from the Artifact
@@ -118,43 +132,55 @@ def create_artifact_from_template(client):
 Wait until the artifact becomes "ready" and then deploy it using task scheduling:
 
 ```python
-def create_task_and_start(client, artifact_id):
-    artifact_manager = client.artifact_manager
+from gmicloud import *
+import time
+from datetime import datetime
 
-    # Wait until the artifact is ready
+def create_task_and_start(client: Client, artifact_id: str) -> str:
+    artifact_manager = client.artifact_manager
+    # Wait for the artifact to be ready
     while True:
-        artifact = artifact_manager.get_artifact(artifact_id)
-        if artifact.build_status == "SUCCESS":
-            break
-        print("Waiting for artifact to be ready...")
+        try:
+            artifact = artifact_manager.get_artifact(artifact_id)
+            print(f"Artifact status: {artifact.build_status}")
+            # Wait until the artifact is ready
+            if artifact.build_status == BuildStatus.SUCCESS:
+                break
+        except Exception as e:
+            raise e
+        # Wait for 2 seconds
         time.sleep(2)
-
-    # Configure and start the task
-    task_manager = client.task_manager
-    task = task_manager.create_task(Task(
-        config=TaskConfig(
-            ray_task_config=RayTaskConfig(
-                ray_version="2.40.0-py310-gpu",
-                file_path="serve",
-                artifact_id=artifact_id,
-                deployment_name="app",
-                replica_resource=ReplicaResource(
-                    cpu=10,
-                    ram_gb=100,
-                    gpu=1,
+    try:
+        task_manager = client.task_manager
+        # Create a task
+        task = task_manager.create_task(Task(
+            config=TaskConfig(
+                ray_task_config=RayTaskConfig(
+                    ray_version="2.40.0-py310-gpu",
+                    file_path="serve",
+                    artifact_id=artifact_id,
+                    deployment_name="app",
+                    replica_resource=ReplicaResource(
+                        cpu=10,
+                        ram_gb=100,
+                        gpu=1,
+                    ),
+                ),
+                task_scheduling=TaskScheduling(
+                    scheduling_oneoff=OneOffScheduling(
+                        trigger_timestamp=int(datetime.now().timestamp()) + 10,
+                        min_replicas=1,
+                        max_replicas=10,
+                    )
                 ),
             ),
-            task_scheduling=TaskScheduling(
-                scheduling_oneoff=OneOffScheduling(
-                    trigger_timestamp=int(datetime.now().timestamp()) + 10,
-                    min_replicas=1,
-                    max_replicas=10,
-                )
-            ),
-        ),
-    ))
+        ))
+
+        # Start the task
+        task_manager.start_task(task.task_id)
+    except Exception as e:
+        raise e
 
-    task_manager.start_task(task.task_id)
     return task.task_id
 ```
 
@@ -163,14 +189,20 @@ def create_task_and_start(client, artifact_id):
 Once the task is running, use the endpoint for inference:
 
 ```python
+from gmicloud import *
 from examples.completion import call_chat_completion
 
-client = Client()
-artifact_id = create_artifact_from_template(client)
-task_id = create_task_and_start(client, artifact_id)
+# Initialize the Client
+cli = Client()
 
-response = call_chat_completion(client, task_id)
-print(response)
+# Create an artifact from a template
+artifact_id = create_artifact_from_template(cli)
+
+# Create a task and start it
+task_id = create_task_and_start(cli, artifact_id)
+
+# Call chat completion
+print(call_chat_completion(cli, task_id))
 ```
 
 ## API Reference
@@ -199,10 +231,10 @@ password: Optional[str] = ""
 
 ## Notes & Troubleshooting
 
-Ensure Credentials are Correct: Double-check your environment variables or parameters passed into the Client object.
-Artifact Status: It may take a few minutes for an artifact or task to transition to the "running" state.
-Inference Endpoint Readiness: Use the task endpoint only after the task status changes to "running".
-Default OpenAI Key: By default, the OpenAI API base URL is derived from the endpoint provided by GMI.
+* Ensure Credentials are Correct: Double-check your environment variables or parameters passed into the Client object.
+* Artifact Status: It may take a few minutes for an artifact or task to transition to the "running" state.
+* Inference Endpoint Readiness: Use the task endpoint only after the task status changes to "running".
+* Default OpenAI Key: By default, the OpenAI API base URL is derived from the endpoint provided by GMI.
 
 ## Contributing
 

{gmicloud-0.1.3 → gmicloud-0.1.5}/README.md

@@ -1,16 +1,11 @@
 # GMICloud SDK (Beta)
 
 ## Overview
-
 Before you start: Our service and GPU resource is currenly invite-only so please contact our team (getstarted@gmicloud.ai) to get invited if you don't have one yet.
 
-The GMI Inference Engine SDK provides a Python interface for deploying and managing machine learning models in
-production environments. It allows users to create model artifacts, schedule tasks for serving models, and call
-inference APIs easily.
+The GMI Inference Engine SDK provides a Python interface for deploying and managing machine learning models in production environments. It allows users to create model artifacts, schedule tasks for serving models, and call inference APIs easily.
 
-This SDK streamlines the process of utilizing GMI Cloud capabilities such as deploying models with Kubernetes-based Ray
-services, managing resources automatically, and accessing model inference endpoints. With minimal setup, developers can
-focus on building ML solutions instead of infrastructure.
+This SDK streamlines the process of utilizing GMI Cloud capabilities such as deploying models with Kubernetes-based Ray services, managing resources automatically, and accessing model inference endpoints. With minimal setup, developers can focus on building ML solutions instead of infrastructure.
 
 ## Features
 
@@ -38,6 +33,7 @@ Set the following environment variables:
 export GMI_CLOUD_CLIENT_ID=<YOUR_CLIENT_ID>
 export GMI_CLOUD_EMAIL=<YOUR_EMAIL>
 export GMI_CLOUD_PASSWORD=<YOUR_PASSWORD>
+export GMI_CLOUD_API_KEY=<YOUR_API_KEY>
 ```
 
 ### Option 2: Passing Credentials as Parameters
@@ -52,7 +48,18 @@ client = Client(client_id="<YOUR_CLIENT_ID>", email="<YOUR_EMAIL>", password="<Y
 
 ## Quick Start
 
-### 1. Create a Task from an Artifact Template
+### 1. How to run the code in the example folder
+```bash
+cd path/to/gmicloud-sdk
+# Create a virtual environment
+python -m venv venv
+source venv/bin/activate
+
+pip install -r requirements.txt
+python -m examples.create_task_from_artifact_template.py
+```
+
+### 2. Create a Task from an Artifact Template
 
 This is the simplest example to deploy an existing artifact template:
 
@@ -81,24 +88,31 @@ response = call_chat_completion(client, task.task_id)
 print(response)
 ```
 
-### 2. Step-by-Step Example: Create Artifact, Task, and Query the Endpoint
+### 3. Step-by-Step Example: Create Artifact, Task, and Query the Endpoint
 
 #### (a) Create an Artifact from a Template
 
 First, you’ll retrieve all templates and create an artifact based on the desired template (e.g., "Llama3.1 8B"):
 
 ```python
-def create_artifact_from_template(client):
+from gmicloud import *
+
+
+def create_artifact_from_template(client: Client) -> str:
     artifact_manager = client.artifact_manager
 
-    # List all available templates
-    templates = artifact_manager.get_artifact_templates()
+    # Get all artifact templates
+    templates = artifact_manager.get_public_templates()
     for template in templates:
         if template.artifact_template_id == "qwen_2.5_14b_instruct_template_001":
-            return artifact_manager.create_artifact_from_template(
-                artifact_template_id=template.artifact_template_id
+            # Create an artifact from a template
+            artifact_id = artifact_manager.create_artifact_from_template(
+                artifact_template_id=template.artifact_template_id,
             )
-    return None
+
+            return artifact_id
+
+    return ""
 ```
 
 #### (b) Create a Task from the Artifact
@@ -106,43 +120,55 @@ def create_artifact_from_template(client):
 Wait until the artifact becomes "ready" and then deploy it using task scheduling:
 
 ```python
-def create_task_and_start(client, artifact_id):
-    artifact_manager = client.artifact_manager
+from gmicloud import *
+import time
+from datetime import datetime
 
-    # Wait until the artifact is ready
+def create_task_and_start(client: Client, artifact_id: str) -> str:
+    artifact_manager = client.artifact_manager
+    # Wait for the artifact to be ready
     while True:
-        artifact = artifact_manager.get_artifact(artifact_id)
-        if artifact.build_status == "SUCCESS":
-            break
-        print("Waiting for artifact to be ready...")
+        try:
+            artifact = artifact_manager.get_artifact(artifact_id)
+            print(f"Artifact status: {artifact.build_status}")
+            # Wait until the artifact is ready
+            if artifact.build_status == BuildStatus.SUCCESS:
+                break
+        except Exception as e:
+            raise e
+        # Wait for 2 seconds
         time.sleep(2)
-
-    # Configure and start the task
-    task_manager = client.task_manager
-    task = task_manager.create_task(Task(
-        config=TaskConfig(
-            ray_task_config=RayTaskConfig(
-                ray_version="2.40.0-py310-gpu",
-                file_path="serve",
-                artifact_id=artifact_id,
-                deployment_name="app",
-                replica_resource=ReplicaResource(
-                    cpu=10,
-                    ram_gb=100,
-                    gpu=1,
+    try:
+        task_manager = client.task_manager
+        # Create a task
+        task = task_manager.create_task(Task(
+            config=TaskConfig(
+                ray_task_config=RayTaskConfig(
+                    ray_version="2.40.0-py310-gpu",
+                    file_path="serve",
+                    artifact_id=artifact_id,
+                    deployment_name="app",
+                    replica_resource=ReplicaResource(
+                        cpu=10,
+                        ram_gb=100,
+                        gpu=1,
+                    ),
+                ),
+                task_scheduling=TaskScheduling(
+                    scheduling_oneoff=OneOffScheduling(
+                        trigger_timestamp=int(datetime.now().timestamp()) + 10,
+                        min_replicas=1,
+                        max_replicas=10,
+                    )
                 ),
             ),
-            task_scheduling=TaskScheduling(
-                scheduling_oneoff=OneOffScheduling(
-                    trigger_timestamp=int(datetime.now().timestamp()) + 10,
-                    min_replicas=1,
-                    max_replicas=10,
-                )
-            ),
-        ),
-    ))
+        ))
+
+        # Start the task
+        task_manager.start_task(task.task_id)
+    except Exception as e:
+        raise e
 
-    task_manager.start_task(task.task_id)
     return task.task_id
 ```
 
@@ -151,14 +177,20 @@ def create_task_and_start(client, artifact_id):
 Once the task is running, use the endpoint for inference:
 
 ```python
+from gmicloud import *
 from examples.completion import call_chat_completion
 
-client = Client()
-artifact_id = create_artifact_from_template(client)
-task_id = create_task_and_start(client, artifact_id)
+# Initialize the Client
+cli = Client()
 
-response = call_chat_completion(client, task_id)
-print(response)
+# Create an artifact from a template
+artifact_id = create_artifact_from_template(cli)
+
+# Create a task and start it
+task_id = create_task_and_start(cli, artifact_id)
+
+# Call chat completion
+print(call_chat_completion(cli, task_id))
 ```
 
 ## API Reference
@@ -187,10 +219,10 @@ password: Optional[str] = ""
 
 ## Notes & Troubleshooting
 
-Ensure Credentials are Correct: Double-check your environment variables or parameters passed into the Client object.
-Artifact Status: It may take a few minutes for an artifact or task to transition to the "running" state.
-Inference Endpoint Readiness: Use the task endpoint only after the task status changes to "running".
-Default OpenAI Key: By default, the OpenAI API base URL is derived from the endpoint provided by GMI.
+* Ensure Credentials are Correct: Double-check your environment variables or parameters passed into the Client object.
+* Artifact Status: It may take a few minutes for an artifact or task to transition to the "running" state.
+* Inference Endpoint Readiness: Use the task endpoint only after the task status changes to "running".
+* Default OpenAI Key: By default, the OpenAI API base URL is derived from the endpoint provided by GMI.
 
 ## Contributing
 

{gmicloud-0.1.3 → gmicloud-0.1.5}/gmicloud/__init__.py

@@ -1,3 +1,6 @@
+import logging
+import os
+
 from ._internal._models import (
     Artifact,
     ArtifactData,
@@ -16,7 +19,8 @@ from ._internal._models import (
 )
 from ._internal._enums import (
     BuildStatus,
-    TaskEndpointStatus
+    TaskEndpointStatus,
+    TaskStatus
 )
 from .client import Client
 
@@ -39,3 +43,10 @@ __all__ = [
     "BuildStatus",
     "TaskEndpointStatus",
 ]
+
+# Configure logging
+log_level = os.getenv("GMI_CLOUD_LOG_LEVEL", "INFO").upper()
+logging.basicConfig(
+    level=log_level,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s'
+)

gmicloud-0.1.5/gmicloud/_internal/_client/_artifact_client.py

@@ -0,0 +1,212 @@
+from typing import List
+import logging
+from requests.exceptions import RequestException
+
+from ._http_client import HTTPClient
+from ._iam_client import IAMClient
+from ._decorator import handle_refresh_token
+from .._models import *
+from .._config import ARTIFACT_SERVICE_BASE_URL
+
+logger = logging.getLogger(__name__)
+
+
+class ArtifactClient:
+    """
+    Client for interacting with the Artifact Service API.
+
+    This client provides methods to perform CRUD operations on artifacts,
+    as well as generating signed URLs for uploading large files.
+    """
+
+    def __init__(self, iam_client: IAMClient):
+        """
+        Initializes the ArtifactClient with an HTTPClient configured
+        to communicate with the Artifact Service base URL.
+        """
+        self.client = HTTPClient(ARTIFACT_SERVICE_BASE_URL)
+        self.iam_client = iam_client
+
+    @handle_refresh_token
+    def get_artifact(self, artifact_id: str) -> Optional[Artifact]:
+        """
+        Fetches an artifact by its ID.
+
+        :param artifact_id: The ID of the artifact to fetch.
+        :return: The Artifact object or None if an error occurs.
+        """
+        try:
+            response = self.client.get(
+                "/get_artifact",
+                self.iam_client.get_custom_headers(),
+                {"artifact_id": artifact_id}
+            )
+            return Artifact.model_validate(response) if response else None
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to fetch artifact {artifact_id}: {e}")
+            return None
+
+    @handle_refresh_token
+    def get_all_artifacts(self) -> List[Artifact]:
+        """
+        Fetches all artifacts.
+
+        :return: A list of Artifact objects. If an error occurs, returns an empty list.
+        """
+        try:
+            response = self.client.get("/get_all_artifacts", self.iam_client.get_custom_headers())
+            if not response:
+                logger.error("Empty response from /get_all_artifacts")
+                return []
+            return [Artifact.model_validate(item) for item in response]
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to fetch all artifacts: {e}")
+            return []
+
+    @handle_refresh_token
+    def create_artifact(self, request: CreateArtifactRequest) -> Optional[CreateArtifactResponse]:
+        """
+        Creates a new artifact in the service.
+
+        :param request: The request object containing artifact details.
+        :return: The response object containing the created artifact details, or None on error.
+        """
+        try:
+            response = self.client.post(
+                "/create_artifact",
+                self.iam_client.get_custom_headers(),
+                request.model_dump()
+            )
+            return CreateArtifactResponse.model_validate(response) if response else None
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to create artifact: {e}")
+            return None
+
+    @handle_refresh_token
+    def create_artifact_from_template(self, artifact_template_id: str) -> Optional[CreateArtifactFromTemplateResponse]:
+        """
+        Creates a new artifact in the service.
+
+        :param artifact_template_id: The ID of the artifact template to use.
+        :return: The response object containing the created artifact details or None if an error occurs.
+        """
+        try:
+            response = self.client.post(
+                "/create_artifact_from_template",
+                self.iam_client.get_custom_headers(),
+                {"artifact_template_id": artifact_template_id}
+            )
+            return CreateArtifactFromTemplateResponse.model_validate(response) if response else None
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to create artifact from template {artifact_template_id}: {e}")
+            return None
+
+    @handle_refresh_token
+    def rebuild_artifact(self, artifact_id: str) -> Optional[RebuildArtifactResponse]:
+        """
+        Rebuilds an artifact in the service.
+
+        :param artifact_id: The ID of the artifact to rebuild.
+        :return: The response object containing the rebuilt artifact details or None if an error occurs.
+        """
+        try:
+            response = self.client.post(
+                "/rebuild_artifact",
+                self.iam_client.get_custom_headers(),
+                {"artifact_id": artifact_id}
+            )
+            return RebuildArtifactResponse.model_validate(response) if response else None
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to rebuild artifact {artifact_id}: {e}")
+            return None
+
+    @handle_refresh_token
+    def delete_artifact(self, artifact_id: str) -> Optional[DeleteArtifactResponse]:
+        """
+        Deletes an artifact by its ID.
+
+        :param artifact_id: The ID of the artifact to delete.
+        :return: The response object containing the deleted artifact details or None if an error occurs.
+        """
+        try:
+            response = self.client.delete(
+                "/delete_artifact",
+                self.iam_client.get_custom_headers(),
+                {"artifact_id": artifact_id}
+            )
+            return DeleteArtifactResponse.model_validate(response) if response else None
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to delete artifact {artifact_id}: {e}")
+            return None
+
+    @handle_refresh_token
+    def get_bigfile_upload_url(self, request: GetBigFileUploadUrlRequest) -> Optional[GetBigFileUploadUrlResponse]:
+        """
+        Generates a pre-signed URL for uploading a large file.
+
+        :param request: The request object containing the artifact ID, file name, and file type.
+        :return: The response object containing the pre-signed URL and upload details, or None if an error occurs.
+        """
+        try:
+            response = self.client.post("/get_bigfile_upload_url",
+                                        self.iam_client.get_custom_headers(),
+                                        request.model_dump())
+
+            if not response:
+                logger.error("Empty response from /get_bigfile_upload_url")
+                return None
+
+            return GetBigFileUploadUrlResponse.model_validate(response)
+
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to generate upload URL: {e}")
+            return None
+
+    @handle_refresh_token
+    def delete_bigfile(self, request: DeleteBigfileRequest) -> Optional[DeleteBigfileResponse]:
+        """
+        Deletes a large file associated with an artifact.
+
+        :param request: The request object containing the artifact ID and file name.
+        :return: The response object containing the deletion status, or None if an error occurs.
+        """
+        try:
+            response = self.client.delete("/delete_bigfile",
+                                          self.iam_client.get_custom_headers(),
+                                          request.model_dump())
+
+            if not response:
+                logger.error("Empty response from /delete_bigfile")
+                return None
+
+            return DeleteBigfileResponse.model_validate(response)
+
+        except (RequestException, ValueError) as e:
+            logger.error(f"Failed to delete big file: {e}")
+            return None
+
+    @handle_refresh_token
+    def get_public_templates(self) -> List[ArtifactTemplate]:
+        """
+        Fetches all artifact templates.
+
+        :return: A list of ArtifactTemplate objects.
+        :rtype: List[ArtifactTemplate]
+        """
+        try:
+            response = self.client.get("/get_public_templates", self.iam_client.get_custom_headers())
+
+            if not response:
+                logger.error("Empty response received from /get_public_templates API")
+                return []
+
+            try:
+                result = GetPublicTemplatesResponse.model_validate(response)
+                return result.artifact_templates
+            except ValueError as ve:
+                logger.error(f"Failed to validate response data: {ve}")
+                return []
+
+        except RequestException as e:
+            logger.error(f"Request to /get_public_templates failed: {e}")
+            return []