synapse-sdk 1.0.0a53__tar.gz → 1.0.0a54__tar.gz

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synapse-sdk
-Version: 1.0.0a53
+Version: 1.0.0a54
 Summary: synapse sdk
 Author-email: datamaker <developer@datamaker.io>
 License: MIT

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk/clients/backend/__init__.py

@@ -1,6 +1,6 @@
 from synapse_sdk.clients.backend.annotation import AnnotationClientMixin
 from synapse_sdk.clients.backend.core import CoreClientMixin
-from synapse_sdk.clients.backend.data_collection import DataCollectionClientMixin
+from synapse_sdk.clients.backend.dataset import DatasetClientMixin
 from synapse_sdk.clients.backend.hitl import HITLClientMixin
 from synapse_sdk.clients.backend.integration import IntegrationClientMixin
 from synapse_sdk.clients.backend.ml import MLClientMixin
@@ -9,7 +9,7 @@ from synapse_sdk.clients.backend.ml import MLClientMixin
 class BackendClient(
     AnnotationClientMixin,
     CoreClientMixin,
-    DataCollectionClientMixin,
+    DatasetClientMixin,
     IntegrationClientMixin,
     MLClientMixin,
     HITLClientMixin,

synapse_sdk-1.0.0a53/synapse_sdk/clients/backend/data_collection.py → synapse_sdk-1.0.0a54/synapse_sdk/clients/backend/dataset.py

@@ -8,18 +8,18 @@ from synapse_sdk.clients.base import BaseClient
 from synapse_sdk.clients.utils import get_batched_list
 
 
-class DataCollectionClientMixin(BaseClient):
-    def list_data_collection(self):
-        path = 'data_collections/'
+class DatasetClientMixin(BaseClient):
+    def list_dataset(self):
+        path = 'datasets/'
         return self._list(path)
 
-    def get_data_collection(self, data_collection_id):
-        """Get data_collection from synapse-backend.
+    def get_dataset(self, dataset_id):
+        """Get dataset from synapse-backend.
 
         Args:
-            data_collection_id: The data_collection id to get.
+            dataset_id: The dataset id to get.
         """
-        path = f'data_collections/{data_collection_id}/?expand=file_specifications'
+        path = f'datasets/{dataset_id}/?expand=file_specifications'
         return self._get(path)
 
     def create_data_file(self, file_path: Path):
@@ -40,19 +40,19 @@ class DataCollectionClientMixin(BaseClient):
         path = 'data_units/'
         return self._post(path, data=data)
 
-    def upload_data_collection(
+    def upload_dataset(
         self,
-        data_collection_id: int,
-        data_collection: Dict,
+        dataset_id: int,
+        dataset: Dict,
         project_id: Optional[int] = None,
         batch_size: int = 1000,
         process_pool: int = 10,
     ):
-        """Upload data_collection to synapse-backend.
+        """Upload dataset to synapse-backend.
 
         Args:
-            data_collection_id: The data_collection id to upload the data to.
-            data_collection: The data_collection to upload.
+            dataset_id: The dataset id to upload the data to.
+            dataset: The dataset to upload.
                 * structure:
                     - files: The files to upload. (key: file name, value: file pathlib object)
                     - meta: The meta data to upload.
@@ -60,14 +60,14 @@ class DataCollectionClientMixin(BaseClient):
             batch_size: The batch size to upload the data.
             process_pool: The process pool to upload the data.
         """
-        # TODO validate data_collection with schema
+        # TODO validate dataset with schema
 
-        params = [(data, data_collection_id) for data in data_collection]
+        params = [(data, dataset_id) for data in dataset]
 
         with Pool(processes=process_pool) as pool:
-            data_collection = pool.starmap(self.upload_data_file, tqdm(params))
+            dataset = pool.starmap(self.upload_data_file, tqdm(params))
 
-        batches = get_batched_list(data_collection, batch_size)
+        batches = get_batched_list(dataset, batch_size)
 
         for batch in tqdm(batches):
             data_units = self.create_data_units(batch)
@@ -82,7 +82,7 @@ class DataCollectionClientMixin(BaseClient):
 
                 self.create_tasks(tasks_data)
 
-    def upload_data_file(self, data: Dict, data_collection_id: int) -> Dict:
+    def upload_data_file(self, data: Dict, dataset_id: int) -> Dict:
         """Upload files to synapse-backend.
 
         Args:
@@ -90,13 +90,13 @@ class DataCollectionClientMixin(BaseClient):
                 * structure:
                     - files: The files to upload. (key: file name, value: file pathlib object)
                     - meta: The meta data to upload.
-            data_collection_id: The data_collection id to upload the data to.
+            dataset_id: The dataset id to upload the data to.
 
         Returns:
             Dict: The result of the upload.
         """
         for name, path in data['files'].items():
             data_file = self.create_data_file(path)
-            data['data_collection'] = data_collection_id
+            data['dataset'] = dataset_id
             data['files'][name] = {'checksum': data_file['checksum'], 'path': str(path)}
         return data

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk/clients/ray/serve.py

@@ -8,6 +8,7 @@ class ServeClientMixin(BaseClient):
         response = self._get(path, params=params)
         for key, item in response['applications'].items():
             response['applications'][key]['deployments'] = list(item['deployments'].values())
+            response['applications'][key]['route_prefix'] = item['route_prefix']
         return list(response['applications'].values())
 
     def get_serve_application(self, pk, params=None):
@@ -15,6 +16,7 @@ class ServeClientMixin(BaseClient):
         response = self._get(path, params=params)
         try:
             response['applications'][pk]['deployments'] = list(response['applications'][pk]['deployments'].values())
+            response['applications'][pk]['route_prefix'] = response['applications'][pk]['route_prefix']
             return response['applications'][pk]
         except KeyError:
             raise ClientError(404, 'Serve Application Not Found')

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk/plugins/categories/export/actions/export.py

@@ -27,13 +27,6 @@ class ExportRun(Run):
         error: str | None = None
         created: str
 
-    class MetricsRecord(BaseModel):
-        """Metrics record model."""
-
-        stand_by: int
-        failed: int
-        success: int
-
     def log_file(
         self, log_type: str, target_id: int, data_file_info: dict, status: ExportStatus, error: str | None = None
     ):
@@ -58,16 +51,6 @@ class ExportRun(Run):
             ).model_dump(),
         )
 
-    def log_metrics(self, record: MetricsRecord, category: str):
-        """Log export metrics.
-
-        Args:
-            record (MetricsRecord): The metrics record to log.
-            category (str): The category of the metrics.
-        """
-        record = self.MetricsRecord.model_validate(record)
-        self.set_metrics(value=record.dict(), category=category)
-
     def export_log_json_file(
         self,
         target_id: int,
@@ -280,7 +263,6 @@ class ExportAction(Action):
             'proportion': 100,
         }
     }
-    metrics_categories = {'data_file', 'original_file'}
 
     def get_filtered_results(self, filters, handler):
         """Get filtered target results."""

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk/plugins/categories/export/templates/plugin/export.py

@@ -42,8 +42,6 @@ def export(run, export_items, path_root, **params):
     origin_files_output_path.mkdir(parents=True, exist_ok=True)
 
     total = params['count']
-    original_file_metrics_record = run.MetricsRecord(stand_by=total, success=0, failed=0)
-    data_file_metrics_record = run.MetricsRecord(stand_by=total, success=0, failed=0)
     # progress init
     run.set_progress(0, total, category='dataset_conversion')
     for no, export_item in enumerate(export_items, start=1):
@@ -58,30 +56,12 @@ def export(run, export_items, path_root, **params):
         if save_original_file_flag:
             if no == 1:
                 run.log_message('Saving original file.')
-            original_status = save_original_file(run, final_data, origin_files_output_path, errors_original_file_list)
-
-            original_file_metrics_record.stand_by -= 1
-            if original_status == ExportStatus.FAILED:
-                original_file_metrics_record.failed += 1
-                continue
-            else:
-                original_file_metrics_record.success += 1
-
-        run.log_metrics(record=original_file_metrics_record, category='original_file')
+            save_original_file(run, final_data, origin_files_output_path, errors_original_file_list)
 
         # Extract data as JSON files
         if no == 1:
             run.log_message('Saving json file.')
-        data_status = save_as_json(run, final_data, json_output_path, errors_json_file_list)
-
-        data_file_metrics_record.stand_by -= 1
-        if data_status == ExportStatus.FAILED:
-            data_file_metrics_record.failed += 1
-            continue
-        else:
-            data_file_metrics_record.success += 1
-
-        run.log_metrics(record=data_file_metrics_record, category='data_file')
+        save_as_json(run, final_data, json_output_path, errors_json_file_list)
 
     run.end_log()
 
@@ -146,7 +126,6 @@ def save_original_file(run, result, base_path, error_file_list):
         status = ExportStatus.FAILED
 
     run.export_log_original_file(result['id'], file_info, status, error_msg)
-    return status
 
 
 def save_as_json(run, result, base_path, error_file_list):
@@ -173,4 +152,3 @@ def save_as_json(run, result, base_path, error_file_list):
         status = ExportStatus.FAILED
 
     run.export_log_json_file(result['id'], file_info, status, error_msg)
-    return status

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk/plugins/categories/upload/actions/upload.py

@@ -105,6 +105,8 @@ class UploadParams(BaseModel):
     storage: int
     collection: int
     project: int | None
+    is_generate_tasks: bool = False
+    is_generate_ground_truths: bool = False
 
     @field_validator('storage', mode='before')
     @classmethod
@@ -129,7 +131,7 @@ class UploadParams(BaseModel):
         action = info.context['action']
         client = action.client
         try:
-            client.get_data_collection(value)
+            client.get_dataset(value)
         except ClientError:
             raise PydanticCustomError('client_error', _('Error occurred while checking collection exists.'))
         return value
@@ -163,6 +165,8 @@ class UploadAction(Action):
         analyze_collection: The progress category for the analyze collection process.
         data_file_upload: The progress category for the upload process.
         generate_data_units: The progress category for the generate data units process.
+        generate_tasks: The progress category for the generate tasks process.
+        generate_ground_truths: The progress category for the generate ground truths process.
     """
 
     name = 'upload'
@@ -171,16 +175,48 @@ class UploadAction(Action):
     run_class = UploadRun
     progress_categories = {
         'analyze_collection': {
-            'proportion': 10,
+            'proportion': 0,
         },
         'upload_data_files': {
-            'proportion': 50,
+            'proportion': 0,
         },
         'generate_data_units': {
-            'proportion': 40,
+            'proportion': 0,
+        },
+        'generate_tasks': {
+            'proportion': 0,
+        },
+        'generate_ground_truths': {
+            'proportion': 0,
         },
     }
 
+    def __init__(self, *args, **kwargs):
+        """Initialize UploadAction."""
+        super().__init__(*args, **kwargs)
+
+        # Setup progress categories ratio by options.
+        progress_ratios = {
+            'upload_only': (5, 60, 35, 0, 0),
+            'generate_tasks': (5, 45, 25, 25, 0),
+            'generate_ground_truths': (5, 35, 30, 15, 15),
+        }
+        options = kwargs['plugin_config']['actions']['upload']['options']
+        progress_categories = self.progress_categories
+        if options['allow_generate_tasks'] and not kwargs['params']['allow_generate_ground_truths']:
+            ratio_name = 'generate_tasks'
+        elif options['allow_generate_ground_truths'] and kwargs['params']['allow_generate_tasks']:
+            ratio_name = 'generate_ground_truths'
+        else:
+            ratio_name = 'upload_only'
+
+        assert len(progress_categories) == len(progress_ratios[ratio_name]), (
+            'Progress categories and ratios length mismatch.'
+        )
+        for i, category in enumerate(progress_categories):
+            progress_categories[category]['proportion'] = progress_ratios[ratio_name][i]
+        self.progress_categories = progress_categories
+
     def get_uploader(self, path):
         """Get uploader from entrypoint."""
         return self.entrypoint(self.run, path)
@@ -229,6 +265,26 @@ class UploadAction(Action):
         generated_data_units = self._generate_data_units(uploaded_files, upload_result_count)
         result['generated_data_units_count'] = len(generated_data_units)
 
+        # Setup task with uploaded synapse-backend data units.
+        if not len(generated_data_units):
+            self.run.log_message('No data units were generated.', context=Context.WARNING.value)
+            self.run.end_log()
+            return result
+
+        if self.config['options']['allow_generate_tasks'] and self.params['is_generate_tasks']:
+            generated_tasks = self._generate_tasks(generated_data_units)
+            result['generated_tasks_count'] = len(generated_tasks)
+        else:
+            self.run.log_message('Generating tasks process has passed.')
+
+        # Generate ground truths for the uploaded data.
+        # TODO: Need to add ground truths generation logic later.
+        if self.config['options']['allow_generate_ground_truths'] and self.params['is_generate_ground_truths']:
+            generated_ground_truths = self._generate_ground_truths()
+            result['generated_ground_truths_count'] = len(generated_ground_truths)
+        else:
+            self.run.log_message('Generating ground truths process has passed.')
+
         self.run.end_log()
         return result
 
@@ -244,7 +300,7 @@ class UploadAction(Action):
 
         client = self.run.client
         collection_id = self.params['collection']
-        collection = client.get_data_collection(collection_id)
+        collection = client.get_dataset(collection_id)
 
         # Finish progress
         self.run.set_progress(1, 1, category='analyze_collection')
@@ -313,3 +369,53 @@ class UploadAction(Action):
         self.run.set_progress(upload_result_count, upload_result_count, category='generate_data_units')
 
         return sum(generated_data_units, [])
+
+    def _generate_tasks(self, generated_data_units: List) -> List:
+        """Setup task with uploaded synapse-backend data units.
+
+        TODO: make batch size configurable.
+        """
+        # Initialize progress
+        self.run.set_progress(0, 1, category='generate_tasks')
+        self.run.log_message('Generating tasks with data files...')
+
+        # Prepare batches for processing
+        client = self.run.client
+        project_id = self.params['project']
+        current_progress = 0
+
+        # Generate tasks
+        generated_tasks = []
+        generated_data_units_count = len(generated_data_units)
+        for data_unit in generated_data_units:
+            tasks_data = []
+            task_data = {'project': project_id, 'data_unit': data_unit['id']}
+            tasks_data.append(task_data)
+            if tasks_data:
+                created_tasks = client.create_tasks(tasks_data)
+                created_task_ids = [created_task['id'] for created_task in created_tasks]
+                generated_tasks.append(created_task_ids)
+                for created_task_id in created_task_ids:
+                    self.run.log_task(created_task_id, UploadStatus.SUCCESS)
+
+            self.run.set_progress(current_progress, generated_data_units_count, category='generate_tasks')
+            current_progress += 1
+
+        # Finish progress
+        self.run.log_message('Generating tasks completed')
+        self.run.set_progress(1, 1, category='generate_tasks')
+
+        return sum(generated_tasks, [])
+
+    def _generate_ground_truths(self):
+        """Generate ground truths for the uploaded data.
+
+        TODO: Need to add ground truths generation logic later.
+        """
+        # Initialize progress
+        self.run.set_progress(0, 1, category='generate_ground_truths')
+        self.run.log_message('Generating ground truths...')
+
+        # Finish progress
+        self.run.log_message('Generating ground truths completed')
+        self.run.set_progress(1, 1, category='generate_ground_truths')

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk/plugins/categories/upload/templates/config.yaml

@@ -1,10 +1,9 @@
 actions:
   upload:
     entrypoint: plugin.upload.Uploader
+    options:
+      allow_generate_tasks: false # Allow the plugin to generate tasks for the uploaded data
+      allow_generate_ground_truths: false # Allow the plugin to generate ground truths for the uploaded data
     supported_data_type: image # A primary data type of synapse backend collection. (e.g. 'image', 'text', 'video', 'pcd', 'audio')
     ui_schema: |
       Dumped FormKit Schema for upload plugin custom options
-  task_pre_annotation:
-    entrypoint: plugin.upload.TaskPreAnnotation
-    ui_schema: |
-      Dumped FormKit Schema for upload plugin custom options

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54/synapse_sdk/plugins/templates/synapse-{{cookiecutter.plugin_code}}-plugin}/.pre-commit-config.yaml

@@ -1,6 +1,6 @@
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.11.10
+    rev: v0.7.1
     hooks:
       - id: ruff
         args: [ --fix ]

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synapse-sdk
-Version: 1.0.0a53
+Version: 1.0.0a54
 Summary: synapse sdk
 Author-email: datamaker <developer@datamaker.io>
 License: MIT

{synapse_sdk-1.0.0a53 → synapse_sdk-1.0.0a54}/synapse_sdk.egg-info/SOURCES.txt

@@ -48,7 +48,7 @@ synapse_sdk/clients/agent/service.py
 synapse_sdk/clients/backend/__init__.py
 synapse_sdk/clients/backend/annotation.py
 synapse_sdk/clients/backend/core.py
-synapse_sdk/clients/backend/data_collection.py
+synapse_sdk/clients/backend/dataset.py
 synapse_sdk/clients/backend/hitl.py
 synapse_sdk/clients/backend/integration.py
 synapse_sdk/clients/backend/ml.py
@@ -117,11 +117,9 @@ synapse_sdk/plugins/categories/smart_tool/templates/plugin/__init__.py
 synapse_sdk/plugins/categories/smart_tool/templates/plugin/auto_label.py
 synapse_sdk/plugins/categories/upload/__init__.py
 synapse_sdk/plugins/categories/upload/actions/__init__.py
-synapse_sdk/plugins/categories/upload/actions/task_pre_annotation.py
 synapse_sdk/plugins/categories/upload/actions/upload.py
 synapse_sdk/plugins/categories/upload/templates/config.yaml
 synapse_sdk/plugins/categories/upload/templates/plugin/__init__.py
-synapse_sdk/plugins/categories/upload/templates/plugin/task_pre_annotation.py
 synapse_sdk/plugins/categories/upload/templates/plugin/upload.py
 synapse_sdk/plugins/templates/cookiecutter.json
 synapse_sdk/plugins/templates/hooks/post_gen_project.py

synapse_sdk-1.0.0a53/synapse_sdk/plugins/categories/upload/actions/task_pre_annotation.py

@@ -1,101 +0,0 @@
-from enum import Enum
-from typing import Annotated
-
-from pydantic import AfterValidator, BaseModel, field_validator
-from pydantic_core import PydanticCustomError
-
-from synapse_sdk.clients.exceptions import ClientError
-from synapse_sdk.plugins.categories.base import Action
-from synapse_sdk.plugins.categories.decorators import register_action
-from synapse_sdk.plugins.enums import PluginCategory, RunMethod
-from synapse_sdk.plugins.models import Run
-from synapse_sdk.utils.pydantic.validators import non_blank
-
-
-class TaskDataAnnotationType(str, Enum):
-    FILE = 'file'
-    INFERENCE = 'inference'
-
-
-class TaskPreAnnotationRun(Run):
-    pass
-
-
-class TaskPreAnnotationParams(BaseModel):
-    """TaskPreAnnotation action parameters.
-
-    Args:
-        name (str): The name of the action.
-        description (str | None): The description of the action.
-        project (int): The project ID.
-        data_collection (int): The data collection ID.
-        task_data_annotation_type (TaskDataAnnotationType): The type of task data annotation.
-    """
-
-    name: Annotated[str, AfterValidator(non_blank)]
-    description: str | None
-    project: int
-    data_collection: int
-    task_data_annotation_type: TaskDataAnnotationType
-
-    @field_validator('data_collection', mode='before')
-    @classmethod
-    def check_data_collection_exists(cls, value: str, info) -> str:
-        """Validate synapse-backend collection exists."""
-        action = info.context['action']
-        client = action.client
-        try:
-            client.get_data_collection(value)
-        except ClientError:
-            raise PydanticCustomError('client_error', 'Error occurred while checking data collection exists.')
-        return value
-
-    @field_validator('project', mode='before')
-    @classmethod
-    def check_project_exists(cls, value: str, info) -> str:
-        """Validate synapse-backend project exists."""
-        if not value:
-            return value
-
-        action = info.context['action']
-        client = action.client
-        try:
-            client.get_project(value)
-        except ClientError:
-            raise PydanticCustomError('client_error', 'Error occurred while checking project exists.')
-        return value
-
-
-@register_action
-class TaskPreAnnotationAction(Action):
-    """TaskPreAnnotation action class.
-
-    * Annotate data to tasks.
-    """
-
-    name = 'task_pre_annotation'
-    category = PluginCategory.UPLOAD
-    method = RunMethod.JOB
-    run_class = TaskPreAnnotationRun
-    progress_categories = {
-        'generate_tasks': {
-            'proportion': 10,
-        },
-        'annotate_task_data': {
-            'proportion': 90,
-        },
-    }
-
-    def start(self):
-        """Start task_pre_annotation action.
-
-        * Generate tasks.
-        * Annotate data to tasks.
-        """
-        task_pre_annotation = self.get_task_pre_annotation()
-        task_pre_annotation.handle_annotate_data_from_files()
-        return {}
-
-    def get_task_pre_annotation(self):
-        """Get task pre annotation entrypoint."""
-        return self.entrypoint()

synapse_sdk-1.0.0a53/synapse_sdk/plugins/categories/upload/templates/plugin/task_pre_annotation.py

@@ -1,14 +0,0 @@
-class TaskPreAnnotation:
-    def __init__(self, run, *args, **kwargs):
-        """Initialize the plugin task pre annotation action class.
-
-        Args:
-            run: Plugin run object.
-        """
-        self.run = run
-
-    def handle_annotate_data_from_files(self):
-        pass
-
-    def handle_annotate_data_with_inference(self):
-        pass