dtlpy 1.86.13__py3-none-any.whl → 1.87.18__py3-none-any.whl

dtlpy/__init__.py

@@ -313,14 +313,14 @@ INSTANCE_CATALOG_REGULAR_XS = InstanceCatalog.REGULAR_XS
 INSTANCE_CATALOG_REGULAR_S = InstanceCatalog.REGULAR_S
 INSTANCE_CATALOG_REGULAR_M = InstanceCatalog.REGULAR_M
 INSTANCE_CATALOG_REGULAR_L = InstanceCatalog.REGULAR_L
-INSTANCE_CATALOG_REGULAR_XL = InstanceCatalog.REGULAR_XL
 INSTANCE_CATALOG_HIGHMEM_XS = InstanceCatalog.HIGHMEM_XS
 INSTANCE_CATALOG_HIGHMEM_S = InstanceCatalog.HIGHMEM_S
 INSTANCE_CATALOG_HIGHMEM_M = InstanceCatalog.HIGHMEM_M
 INSTANCE_CATALOG_HIGHMEM_L = InstanceCatalog.HIGHMEM_L
-INSTANCE_CATALOG_HIGHMEM_XL = InstanceCatalog.HIGHMEM_XL
 INSTANCE_CATALOG_GPU_K80_S = InstanceCatalog.GPU_K80_S
 INSTANCE_CATALOG_GPU_K80_M = InstanceCatalog.GPU_K80_M
+INSTANCE_CATALOG_GPU_T4_S = InstanceCatalog.GPU_T4_S
+INSTANCE_CATALOG_GPU_T4_M = InstanceCatalog.GPU_T4_M
 
 MODALITY_TYPE_OVERLAY = ModalityTypeEnum.OVERLAY
 MODALITY_TYPE_PREVIEW = ModalityTypeEnum.PREVIEW

dtlpy/__version__.py

@@ -1 +1 @@
-version = '1.86.13'
+version = '1.87.18'

dtlpy/entities/annotation.py

@@ -235,9 +235,7 @@ class Annotation(entities.BaseEntity):
     def coordinates(self):
         color = None
         if self.type in ['binary']:
-            color = self.annotation_definition._color
-            if color is None:
-                color = self.color
+            color = self.color
         coordinates = self.annotation_definition.to_coordinates(color=color)
         return coordinates
 
@@ -446,16 +444,16 @@ class Annotation(entities.BaseEntity):
     @property
     def color(self):
         # if "dataset" is not in self - this will always get the dataset
-        if self.type == 'binary' and self.annotation_definition._color is not None:
-            color = self.annotation_definition._color
+        try:
+            colors = self.dataset._get_ontology().color_map
+        except (exceptions.BadRequest, exceptions.NotFound):
+            colors = None
+            logger.warning('Cant get dataset for annotation color. using default.')
+        if colors is not None and self.label in colors:
+            color = colors[self.label]
         else:
-            try:
-                colors = self.dataset._get_ontology().color_map
-            except (exceptions.BadRequest, exceptions.NotFound):
-                colors = None
-                logger.warning('Cant get dataset for annotation color. using default.')
-            if colors is not None and self.label in colors:
-                color = colors[self.label]
+            if self.type == 'binary' and self.annotation_definition._color is not None:
+                color = self.annotation_definition._color
             else:
                 color = (255, 255, 255)
         return color

dtlpy/entities/annotation_collection.py

@@ -94,8 +94,6 @@ class AnnotationCollection(entities.BaseEntity):
                                          'name': model_info.get('name'),
                                          'model_id': model_info.get('model_id'),
                                          }
-            metadata['user']['annotation_type'] = 'prediction'
-
         if prompt_id is not None:
             if metadata is None:
                 metadata = dict()

dtlpy/entities/command.py

@@ -141,7 +141,7 @@ class Command(entities.BaseEntity):
                                entities.CommandsStatus.FINALIZING,
                                entities.CommandsStatus.IN_PROGRESS]
 
-    def wait(self, timeout=0, step=None,  backoff_factor=0.1):
+    def wait(self, timeout=0, step=None,  backoff_factor=1):
         """
         Wait for Command to finish
 

dtlpy/entities/dataset.py

@@ -244,9 +244,8 @@ class Dataset(entities.BaseEntity):
 
     @readonly.setter
     def readonly(self, state):
-        raise exceptions.PlatformException(
-            error='400',
-            message='Cannot set attribute readonly. Please use "set_readonly({})" method'.format(state))
+        import warnings
+        warnings.warn("`readonly` flag on dataset is deprecated, doing nothing.", DeprecationWarning)
 
     @property
     def labels_flat_dict(self):
@@ -518,11 +517,8 @@ class Dataset(entities.BaseEntity):
 
             dataset.set_readonly(state=True)
         """
-        if not isinstance(state, bool):
-            raise exceptions.PlatformException(
-                error='400',
-                message='Argument "state" must be bool. input type: {}'.format(type(state)))
-        return self.datasets.set_readonly(dataset=self, state=state)
+        import warnings
+        warnings.warn("`readonly` flag on dataset is deprecated, doing nothing.", DeprecationWarning)
 
     def clone(self,
               clone_name=None,

dtlpy/entities/filters.py

@@ -33,7 +33,7 @@ class FiltersResource(str, Enum):
     MODEL = "models"
     WEBHOOK = "webhooks"
     RECIPE = 'recipe'
-    DATASET = 'dataset'
+    DATASET = 'datasets'
     ONTOLOGY = 'ontology'
     TASK = 'tasks'
     PIPELINE = 'pipeline'
@@ -113,6 +113,7 @@ class Filters:
         self._ref_op = None
         self._ref_assignment_id = None
         self._ref_task_id = None
+        self._system_space = None
 
         self._use_defaults = use_defaults
         self.__add_defaults()
@@ -146,6 +147,14 @@ class Filters:
         self.reset()
         self.__add_defaults()
 
+    @property
+    def system_space(self):
+        return self._system_space
+
+    @system_space.setter
+    def system_space(self, val: bool):
+        self._system_space = val
+
     def reset(self):
         self.or_filter_list = list()
         self.and_filter_list = list()
@@ -443,7 +452,8 @@ class Filters:
 
         if self.context is not None:
             _json['context'] = self.context
-
+        if self._system_space is not None:
+            _json['systemSpace'] = self._system_space
         return _json
 
     def sort_by(self, field, value: FiltersOrderByDirection = FiltersOrderByDirection.ASCENDING):

dtlpy/entities/model.py

@@ -105,6 +105,7 @@ class Model(entities.BaseEntity):
     _client_api = attr.ib(type=ApiClient, repr=False)
     _repositories = attr.ib(repr=False)
     _ontology = attr.ib(repr=False, default=None)
+    updated_by = attr.ib(default=None)
 
     @staticmethod
     def _protected_from_json(_json, client_api, project, package, is_fetched=True):
@@ -184,7 +185,8 @@ class Model(entities.BaseEntity):
             context=_json.get('context', {}),
             input_type=_json.get('inputType', None),
             output_type=_json.get('outputType', None),
-            module_name = _json.get('moduleName', None)
+            module_name=_json.get('moduleName', None),
+            updated_by=_json.get('updatedBy', None)
         )
         inst.is_fetched = is_fetched
         return inst
@@ -212,6 +214,7 @@ class Model(entities.BaseEntity):
                                                         attr.fields(Model).updated_at,
                                                         attr.fields(Model).input_type,
                                                         attr.fields(Model).output_type,
+                                                        attr.fields(Model).updated_by
                                                         ))
         _json['packageId'] = self.package_id
         _json['datasetId'] = self.dataset_id
@@ -230,6 +233,10 @@ class Model(entities.BaseEntity):
                 artifact = artifact.to_json(as_artifact=True)
             model_artifacts.append(artifact)
         _json['artifacts'] = model_artifacts
+
+        if self.updated_by:
+            _json['updatedBy'] = self.updated_by
+
         return _json
 
     ############
@@ -307,7 +314,9 @@ class Model(entities.BaseEntity):
                  dpks=repositories.Dpks(client_api=self._client_api),
                  services=repositories.Services(client_api=self._client_api,
                                                 project=self._project,
-                                                project_id=self.project_id),
+                                                project_id=self.project_id,
+                                                model_id=self.id,
+                                                model=self),
                  )
         return r
 

dtlpy/entities/node.py

@@ -94,7 +94,8 @@ class PipelineNodeIO:
                  action: str = None,
                  default_value=None,
                  variable_name: str = None,
-                 actions: list = None):
+                 actions: list = None,
+                 description: str = None):
         """
         Pipeline Node
 
@@ -116,6 +117,7 @@ class PipelineNodeIO:
         self.port_percentage = port_percentage
         self.default_value = default_value
         self.variable_name = variable_name
+        self.description = description
 
         if action is not None:
             warnings.warn('action param has been deprecated in version 1.80', DeprecationWarning)
@@ -141,6 +143,7 @@ class PipelineNodeIO:
             default_value=_json.get('defaultValue', None),
             variable_name=_json.get('variableName', None),
             actions=_json.get('actions', None),
+            description=_json.get('description', None),
         )
 
     def to_json(self):
@@ -158,7 +161,8 @@ class PipelineNodeIO:
             _json['actions'] = self.actions
         if self.default_value:
             _json['defaultValue'] = self.default_value
-
+        if self.description:
+            _json['description'] = self.description
         return _json
 
 

dtlpy/entities/ontology.py

@@ -396,9 +396,9 @@ class Ontology(entities.BaseEntity):
     def _add_image_label(self, icon_path):
         display_data = dict()
         if self.project is not None:
-            dataset = self.project.datasets.get(dataset_name='Binaries')
+            dataset = self.project.datasets._get_binaries_dataset()
         elif self.dataset is not None:
-            dataset = self.dataset.project.datasets.get(dataset_name='Binaries')
+            dataset = self.dataset.project.datasets._get_binaries_dataset()
         else:
             raise ValueError('must have project or dataset to create with icon path')
         platform_path = "/.dataloop/ontologies/{}/labelDisplayImages/".format(self.id)

dtlpy/entities/package_function.py

@@ -95,6 +95,7 @@ class FunctionIO(entities.DlEntity):
     value = entities.DlProperty(location=['value'], _type=str)
     name = entities.DlProperty(location=['name'], _type=str)
     actions = entities.DlProperty(location=['actions'], _type=list)
+    description = entities.DlProperty(location=['description'], _type=str)
 
     def __repr__(self):
         # TODO need to move to DlEntity
@@ -164,6 +165,8 @@ class FunctionIO(entities.DlEntity):
             }
             if self.actions:
                 _json['actions'] = self.actions
+            if self.description:
+                _json['description'] = self.description
         elif resource in ['execution', 'service']:
             _json = {
                 self.name: self.value

dtlpy/entities/pipeline.py

@@ -229,6 +229,8 @@ class Pipeline(entities.BaseEntity):
     _original_variables = attr.ib(repr=False, type=List[Variable])
     _repositories = attr.ib(repr=False)
 
+    updated_by = attr.ib(default=None)
+
     @staticmethod
     def _protected_from_json(_json, client_api, project, is_fetched=True):
         """
@@ -298,6 +300,7 @@ class Pipeline(entities.BaseEntity):
             status=_json.get('status', None),
             original_settings=settings,
             original_variables=json_variables,
+            updated_by=_json.get('updatedBy', None),
         )
         for node in _json.get('nodes', list()):
             inst.nodes.add(node=cls.pipeline_node(node))
@@ -352,7 +355,8 @@ class Pipeline(entities.BaseEntity):
                                                         attr.fields(Pipeline).settings,
                                                         attr.fields(Pipeline).variables,
                                                         attr.fields(Pipeline)._original_settings,
-                                                        attr.fields(Pipeline)._original_variables
+                                                        attr.fields(Pipeline)._original_variables,
+                                                        attr.fields(Pipeline).updated_by,
                                                         ))
 
         _json['projectId'] = self.project_id
@@ -377,6 +381,8 @@ class Pipeline(entities.BaseEntity):
             _json['description'] = self.description
         if self.revisions is not None:
             _json['revisions'] = self.revisions
+        if self.updated_by is not None:
+            _json['updatedBy'] = self.updated_by
 
         return _json
 

dtlpy/entities/recipe.py

@@ -252,7 +252,7 @@ class Recipe(entities.BaseEntity):
                                                message='file Must be pdf')
         for project_id in self.project_ids:
             project = repositories.Projects(client_api=self._client_api).get(project_id=project_id)
-            dataset = project.datasets.get(dataset_name='Binaries')
+            dataset = project.datasets._get_binaries_dataset()
             remote_path = '/.dataloop/recipes/{}/instructions'.format(self.id)
             instruction_item = dataset.items.upload(local_path=annotation_instruction_file,
                                                     remote_path=remote_path,

dtlpy/entities/service.py

@@ -80,8 +80,6 @@ class InstanceCatalog(str, Enum):
          - regular pod with medium size
        * - REGULAR_L
          - regular pod with large size
-       * - REGULAR_XL
-         - regular pod with extra large size
        * - HIGHMEM_XS
          - highmem pod with extra small size
        * - HIGHMEM_S
@@ -90,25 +88,27 @@ class InstanceCatalog(str, Enum):
          - highmem pod with medium size
        * - HIGHMEM_L
          - highmem pod with large size
-       * - HIGHMEM_XL
-         - highmem pod with extra large size
        * - GPU_K80_S
-         - GPU pod with small size
+         - GPU NVIDIA K80 pod with small size
        * - GPU_K80_M
-         - GPU pod with medium size
+         - GPU NVIDIA K80 pod with medium size
+       * - GPU_T4_S
+         - GPU NVIDIA T4 pod with regular memory
+       * - GPU_T4_M
+         - GPU NVIDIA T4 pod with highmem
     """
     REGULAR_XS = "regular-xs"
     REGULAR_S = "regular-s"
     REGULAR_M = "regular-m"
     REGULAR_L = "regular-l"
-    REGULAR_XL = "regular-xl"
     HIGHMEM_XS = "highmem-xs"
     HIGHMEM_S = "highmem-s"
     HIGHMEM_M = "highmem-m"
     HIGHMEM_L = "highmem-l"
-    HIGHMEM_XL = "highmem-xl"
     GPU_K80_S = "gpu-k80-s"
     GPU_K80_M = "gpu-k80-m"
+    GPU_T4_S = "gpu-t4"
+    GPU_T4_M = "gpu-t4-m"
 
 
 class RuntimeType(str, Enum):
@@ -223,6 +223,7 @@ class Service(entities.BaseEntity):
     max_attempts = attr.ib()
     mode = attr.ib(repr=False)
     metadata = attr.ib()
+    archive = attr.ib(repr=False)
 
     # SDK
     _package = attr.ib(repr=False)
@@ -231,6 +232,7 @@ class Service(entities.BaseEntity):
     # repositories
     _project = attr.ib(default=None, repr=False)
     _repositories = attr.ib(repr=False)
+    updated_by = attr.ib(default=None)
 
     @property
     def createdAt(self):
@@ -265,7 +267,7 @@ class Service(entities.BaseEntity):
         return status, service
 
     @classmethod
-    def from_json(cls, _json: dict, client_api: ApiClient=None, package=None, project=None, is_fetched=True):
+    def from_json(cls, _json: dict, client_api: ApiClient = None, package=None, project=None, is_fetched=True):
         """
         Build a service entity object from a json
 
@@ -325,7 +327,9 @@ class Service(entities.BaseEntity):
             secrets=_json.get("secrets", None),
             type=_json.get("type", None),
             mode=_json.get('mode', dict()),
-            metadata=_json.get('metadata', None)
+            metadata=_json.get('metadata', None),
+            archive=_json.get('archive', None),
+            updated_by=_json.get('updatedBy', None)
         )
         inst.is_fetched = is_fetched
         return inst
@@ -449,7 +453,9 @@ class Service(entities.BaseEntity):
                 attr.fields(Service).secrets,
                 attr.fields(Service)._type,
                 attr.fields(Service).mode,
-                attr.fields(Service).metadata
+                attr.fields(Service).metadata,
+                attr.fields(Service).archive,
+                attr.fields(Service).updated_by,
             )
         )
 
@@ -469,6 +475,9 @@ class Service(entities.BaseEntity):
         _json['createdAt'] = self.created_at
         _json['updatedAt'] = self.updated_at
 
+        if self.updated_by is not None:
+            _json['updatedBy'] = self.updated_by
+
         if self.max_attempts is not None:
             _json['maxAttempts'] = self.max_attempts
 
@@ -493,6 +502,8 @@ class Service(entities.BaseEntity):
         if self.metadata:
             _json['metadata'] = self.metadata
 
+        if self.archive:
+            _json['archive'] = self.archive
         return _json
 
     def update(self, force=False):

dtlpy/entities/task.py

@@ -92,6 +92,7 @@ class Task:
     available_actions = attr.ib()
     total_items = attr.ib()
     priority = attr.ib()
+    _description = attr.ib()
 
     # sdk
     _client_api = attr.ib(repr=False)
@@ -102,6 +103,18 @@ class Task:
     _tasks = attr.ib(default=None, repr=False)
     _settings = attr.ib(default=None, repr=False)
 
+    @property
+    def description(self):
+        return self._description
+
+    @description.setter
+    def description(self, description):
+        if not isinstance(description, str):
+            raise ValueError('description should be a string')
+        if self._description is None:
+            self._description = {}
+        self._description['content'] = description
+
     @staticmethod
     def _protected_from_json(_json, client_api, project, dataset):
         """
@@ -176,7 +189,8 @@ class Task:
             created_at=_json.get('createdAt', None),
             available_actions=actions,
             total_items=_json.get('totalItems', None),
-            priority=_json.get('priority', None)
+            priority=_json.get('priority', None),
+            description=_json.get('description', None)
         )
 
     def to_json(self):
@@ -208,6 +222,7 @@ class Task:
                 attr.fields(Task).created_at,
                 attr.fields(Task).total_items,
                 attr.fields(Task)._settings,
+                attr.fields(Task)._description
             )
         )
         _json['projectId'] = self.project_id
@@ -217,6 +232,7 @@ class Task:
         _json['dueDate'] = self.due_date
         _json['totalItems'] = self.total_items
         _json['forReview'] = self.for_review
+        _json['description'] = self.description
 
         if self.available_actions is not None:
             _json['availableActions'] = [action.to_json() for action in self.available_actions]
@@ -472,3 +488,4 @@ class Task:
         :rtype: bool
         """
         return self.tasks.set_status(status=status, operation=operation, item_ids=item_ids, task_id=self.id)
+    

dtlpy/entities/trigger.py

@@ -84,6 +84,8 @@ class BaseTrigger(entities.BaseEntity):
     _op_type = attr.ib(default='service')
     _repositories = attr.ib(repr=False)
 
+    updated_by = attr.ib(default=None)
+
     @staticmethod
     def _get_operation(operation):
         op_type = operation.get('type', None)
@@ -243,6 +245,7 @@ class BaseTrigger(entities.BaseEntity):
                                                               attr.fields(BaseTrigger).created_at,
                                                               attr.fields(BaseTrigger).updated_at,
                                                               attr.fields(BaseTrigger).operation,
+                                                              attr.fields(BaseTrigger).updated_by,
                                                               ))
 
         # rename
@@ -251,6 +254,8 @@ class BaseTrigger(entities.BaseEntity):
         _json['updatedAt'] = self.updated_at
         if self.is_global is not None:
             _json['global'] = self.is_global
+        if self.updated_by is not None:
+            _json['updatedBy'] = self.updated_by
         return _json
 
     def delete(self):
@@ -342,7 +347,8 @@ class Trigger(BaseTrigger):
             op_type=operation.get('type', None),
             spec=spec,
             pipeline_id=pipeline_id,
-            operation=operation
+            operation=operation,
+            updated_by=_json.get('updatedBy', None),
         )
 
 

dtlpy/ml/base_model_adapter.py

@@ -1,10 +1,10 @@
-import copy
 import tempfile
 import datetime
 import logging
 import shutil
 import base64
 import tqdm
+import sys
 import io
 import os
 from PIL import Image
@@ -309,8 +309,6 @@ class BaseModelAdapter(utilities.BaseServiceRunner):
         if cleanup:
             shutil.rmtree(path=local_path, ignore_errors=True)
             self.logger.info("Clean-up. deleting {}".format(local_path))
-        self.model_entity.status = 'trained'
-        self.model_entity = self.model_entity.update()
 
     # ===============
     # SERVICE METHODS
@@ -339,7 +337,7 @@ class BaseModelAdapter(utilities.BaseServiceRunner):
         pool = ThreadPoolExecutor(max_workers=16)
 
         annotations = list()
-        for i_batch in tqdm.tqdm(range(0, len(items), batch_size), desc='predicting', unit='bt', leave=None):
+        for i_batch in tqdm.tqdm(range(0, len(items), batch_size), desc='predicting', unit='bt', leave=None, file=sys.stdout):
             batch_items = items[i_batch: i_batch + batch_size]
             batch = list(pool.map(self.prepare_item_func, batch_items))
             batch_collections = self.predict(batch, **kwargs)
@@ -415,12 +413,11 @@ class BaseModelAdapter(utilities.BaseServiceRunner):
                     cleanup=False,
                     progress: utilities.Progress = None,
                     context: utilities.Context = None):
-        # FROM PARENT
         """
-            Train on existing model.
-            data will be taken from dl.Model.datasetId
-            configuration is as defined in dl.Model.configuration
-            upload the output the model's bucket (model.bucket)
+        Train on existing model.
+        data will be taken from dl.Model.datasetId
+        configuration is as defined in dl.Model.configuration
+        upload the output the model's bucket (model.bucket)
         """
         if isinstance(model, dict):
             model = repositories.Models(client_api=self._client_api).get(model_id=model['id'])
@@ -469,7 +466,8 @@ class BaseModelAdapter(utilities.BaseServiceRunner):
                                 progress=99)
 
             self.save_to_model(local_path=output_path, replace=True)
-
+            model.status = 'trained'
+            model.update()
             ###########
             # cleanup #
             ###########

dtlpy/ml/train_utils.py

@@ -58,4 +58,3 @@ def prepare_dataset(dataset: entities.Dataset,
         cloned_dataset.metadata['system']['clone_info'].update({'filters': json.dumps(filters.prepare())})
     cloned_dataset.update(system_metadata=True)
     return cloned_dataset
-    # cloned_dataset.set_readonly(True)

dtlpy/new_instance.py

@@ -166,14 +166,16 @@ class Dtlpy:
         REGULAR_S = 'regular-s'
         REGULAR_M = 'regular-m'
         REGULAR_L = 'regular-l'
-        REGULAR_XL = 'regular-xl'
         HIGHMEM_MICRO = 'highmem-micro'
         HIGHMEM_XS = 'highmem-xs'
         HIGHMEM_S = 'highmem-s'
         HIGHMEM_M = 'highmem-m'
         HIGHMEM_L = 'highmem-l'
-        HIGHMEM_XL = 'highmem-xl'
-        GPU_K80_S = 'gpu-k80-s'
+        GPU_K80_S = "gpu-k80-s"
+        GPU_K80_M = "gpu-k80-m"
+        GPU_T4_S = "gpu-t4-s"
+        GPU_T4_M = "gpu-t4-m"
+
 
     class LoggingLevel:
         DEBUG = 'debug'

dtlpy/repositories/artifacts.py

@@ -22,7 +22,7 @@ class Artifacts:
                  project_id: str = None,
                  model: entities.Model = None,
                  package: entities.Package = None,
-                 dataset_name='Binaries'):
+                 dataset_name=None):
         self._client_api = client_api
         self._project = project
         self._dataset = dataset
@@ -40,21 +40,15 @@ class Artifacts:
         if self._dataset is None:
             # get dataset from project
             try:
-                self._dataset = self.project.datasets.get(dataset_name=self.dataset_name)
+                if self.dataset_name is None:
+                    self.dataset_name = 'Binaries'
+                    self._dataset = self.project.datasets._get_binaries_dataset()
+                else:
+                    self._dataset = self.project.datasets.get(dataset_name=self.dataset_name)
             except exceptions.NotFound:
-                self._dataset = None
-            if self._dataset is None:
-                logger.debug(
-                    'Dataset for artifacts was not found. Creating... dataset name: {ds!r}. project_id={id}'.format(
-                        ds=self.dataset_name, id=self.project.id))
-                self._dataset = self.project.datasets.create(dataset_name=self.dataset_name)
-                # add system to metadata
-                if 'metadata' not in self._dataset.to_json():
-                    self._dataset.metadata = dict()
-                if 'system' not in self._dataset.metadata:
-                    self._dataset.metadata['system'] = dict()
-                self._dataset.metadata['system']['scope'] = 'system'
-                self.project.datasets.update(dataset=self._dataset, system_metadata=True)
+                raise ValueError(
+                    f'Missing "{self.dataset_name}" dataset in the project. Please contact support for help')
+
         return self._dataset
 
     @property

dtlpy/repositories/codebases.py

@@ -65,21 +65,9 @@ class Codebases:
         if self._dataset is None:
             # get dataset from project
             try:
-                self._dataset = self.project.datasets.get(dataset_name='Binaries')
+                self._dataset = self.project.datasets._get_binaries_dataset()
             except exceptions.NotFound:
-                self._dataset = None
-            if self._dataset is None:
-                logger.debug(
-                    'Dataset for codebases was not found. Creating... dataset name: "Binaries". project_id={}'.format(
-                        self.project.id))
-                self._dataset = self.project.datasets.create(dataset_name='Binaries')
-                # add system to metadata
-                if 'metadata' not in self._dataset.to_json():
-                    self._dataset.metadata = dict()
-                if 'system' not in self._dataset.metadata:
-                    self._dataset.metadata['system'] = dict()
-                self._dataset.metadata['system']['scope'] = 'system'
-                self.project.datasets.update(dataset=self._dataset, system_metadata=True)
+                raise ValueError('Missing "Binaries" dataset in the project. Please contact support for help')
         assert isinstance(self._dataset, entities.Dataset)
         return self._dataset