clarifai 10.0.1__py3-none-any.whl → 10.1.1__py3-none-any.whl

clarifai/client/model.py

@@ -1,6 +1,6 @@
 import os
 import time
-from typing import Any, Dict, Generator, List
+from typing import Any, Dict, Generator, List, Union
 
 import requests
 import yaml
@@ -32,6 +32,7 @@ class Model(Lister, BaseClient):
                model_version: Dict = {'id': ""},
                base_url: str = "https://api.clarifai.com",
                pat: str = None,
+               token: str = None,
                **kwargs):
     """Initializes a Model object.
 
@@ -41,6 +42,7 @@ class Model(Lister, BaseClient):
         model_version (dict): The Model Version to interact with.
         base_url (str): Base API url. Default "https://api.clarifai.com"
         pat (str): A personal access token for authentication. Can be set as env var CLARIFAI_PAT
+        token (str): A session token for authentication. Accepts either a session token or a pat. Can be set as env var CLARIFAI_SESSION_TOKEN
         **kwargs: Additional keyword arguments to be passed to the Model.
     """
     if url and model_id:
@@ -53,9 +55,10 @@ class Model(Lister, BaseClient):
       kwargs = {'user_id': user_id, 'app_id': app_id}
     self.kwargs = {**kwargs, 'id': model_id, 'model_version': model_version,}
     self.model_info = resources_pb2.Model(**self.kwargs)
-    self.logger = get_logger(logger_level="INFO")
+    self.logger = get_logger(logger_level="INFO", name=__name__)
     self.training_params = {}
-    BaseClient.__init__(self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat)
+    BaseClient.__init__(
+        self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat, token=token)
     Lister.__init__(self)
 
   def list_training_templates(self) -> List[str]:
@@ -212,6 +215,8 @@ class Model(Lister, BaseClient):
         >>> model_params = model.get_params(template='template', yaml_file='model_params.yaml')
         >>> model.train('model_params.yaml')
     """
+    if not self.model_info.model_type_id:
+      self.load_info()
     if self.model_info.model_type_id not in TRAINABLE_MODEL_TYPES:
       raise UserError(f"Model type {self.model_info.model_type_id} is not trainable")
     if not yaml_file and len(self.training_params) == 0:
@@ -222,8 +227,10 @@ class Model(Lister, BaseClient):
         params_dict = yaml.safe_load(file)
     else:
       params_dict = self.training_params
-
-    train_dict = params_parser(params_dict)
+    #getting all the concepts for the model type
+    if self.model_info.model_type_id not in ["clusterer", "text-to-text"]:
+      concepts = self._list_concepts()
+    train_dict = params_parser(params_dict, concepts)
     request = service_pb2.PostModelVersionsRequest(
         user_app_id=self.user_app_id,
         model_id=self.id,
@@ -331,7 +338,7 @@ class Model(Lister, BaseClient):
     dict_response = MessageToDict(response, preserving_proto_field_name=True)
     kwargs = self.process_response_keys(dict_response['model'], 'model')
 
-    return Model(base_url=self.base, pat=self.pat, **kwargs)
+    return Model(base_url=self.base, pat=self.pat, token=self.token, **kwargs)
 
   def list_versions(self, page_no: int = None,
                     per_page: int = None) -> Generator['Model', None, None]:
@@ -373,11 +380,8 @@ class Model(Lister, BaseClient):
         del model_version_info['train_info']['dataset']['version']['metrics']
       except KeyError:
         pass
-      yield Model(
-          model_id=self.id,
-          base_url=self.base,
-          pat=self.pat,
-          **dict(self.kwargs, model_version=model_version_info))
+      yield Model.from_auth_helper(
+          model_id=self.id, **dict(self.kwargs, model_version=model_version_info))
 
   def predict(self, inputs: List[Input], inference_params: Dict = {}, output_config: Dict = {}):
     """Predicts the model based on the given inputs.
@@ -548,6 +552,17 @@ class Model(Lister, BaseClient):
         resources_pb2.OutputInfo(
             output_config=resources_pb2.OutputConfig(**output_config), params=params))
 
+  def _list_concepts(self) -> List[str]:
+    """Lists all the concepts for the model type.
+
+    Returns:
+        concepts (List): List of concepts for the model type.
+    """
+    request_data = dict(user_app_id=self.user_app_id)
+    all_concepts_infos = self.list_pages_generator(self.STUB.ListConcepts,
+                                                   service_pb2.ListConceptsRequest, request_data)
+    return [concept_info['concept_id'] for concept_info in all_concepts_infos]
+
   def load_info(self) -> None:
     """Loads the model info."""
     request = service_pb2.GetModelRequest(
@@ -576,3 +591,169 @@ class Model(Lister, BaseClient):
         if hasattr(self.model_info, param)
     ]
     return f"Model Details: \n{', '.join(attribute_strings)}\n"
+
+  def list_evaluations(self) -> resources_pb2.EvalMetrics:
+    """List all eval_metrics of current model version
+
+    Raises:
+        Exception: Failed to call API
+
+    Returns:
+        resources_pb2.EvalMetrics
+    """
+    assert self.model_info.model_version.id, "Model version is empty. Please provide `model_version` as arguments or with a URL as the format '{user_id}/{app_id}/models/{your_model_id}/model_version_id/{your_version_model_id}' when initializing."
+    request = service_pb2.ListModelVersionEvaluationsRequest(
+        user_app_id=self.user_app_id,
+        model_id=self.id,
+        model_version_id=self.model_info.model_version.id)
+    response = self._grpc_request(self.STUB.ListModelVersionEvaluations, request)
+
+    if response.status.code != status_code_pb2.SUCCESS:
+      raise Exception(response.status)
+
+    return response.eval_metrics
+
+  def evaluate(self,
+               dataset_id: str,
+               dataset_app_id: str = None,
+               dataset_user_id: str = None,
+               eval_id: str = None,
+               extended_metrics: dict = None,
+               eval_info: dict = None) -> resources_pb2.EvalMetrics:
+    """ Run evaluation
+
+    Args:
+      dataset_id (str): Dataset Id.
+      dataset_app_id (str): App ID for cross app evaluation, leave it as None to use Model App ID. Default is None.
+      dataset_user_id (str): User ID for cross app evaluation, leave it as None to use Model User ID. Default is None.
+      eval_id (str): Specific ID for the evaluation. You must specify this parameter to either overwrite the result with the dataset ID or format your evaluation in an informative manner. If you don't, it will use random ID from system. Default is None.
+      extended_metrics (dict): user custom metrics result. Default is None.
+      eval_info (dict): custom eval info. Default is empty dict.
+
+    Return
+      eval_metrics
+
+    """
+    assert self.model_info.model_version.id, "Model version is empty. Please provide `model_version` as arguments or with a URL as the format '{user_id}/{app_id}/models/{your_model_id}/model_version_id/{your_version_model_id}' when initializing."
+    metrics = None
+    if isinstance(extended_metrics, dict):
+      metrics = Struct()
+      metrics.update(extended_metrics)
+      metrics = resources_pb2.ExtendedMetrics(user_metrics=metrics)
+
+    eval_info_params = None
+    if isinstance(eval_info, dict):
+      eval_info_params = Struct()
+      eval_info_params.update(eval_info)
+      eval_info_params = resources_pb2.EvalInfo(params=eval_info_params)
+
+    eval_metric = resources_pb2.EvalMetrics(
+        id=eval_id,
+        model=resources_pb2.Model(
+            id=self.id,
+            app_id=self.auth_helper.app_id,
+            user_id=self.auth_helper.user_id,
+            model_version=resources_pb2.ModelVersion(id=self.model_info.model_version.id),
+        ),
+        extended_metrics=metrics,
+        ground_truth_dataset=resources_pb2.Dataset(
+            id=dataset_id,
+            app_id=dataset_app_id or self.auth_helper.app_id,
+            user_id=dataset_user_id or self.auth_helper.user_id,
+        ),
+        eval_info=eval_info_params,
+    )
+    request = service_pb2.PostEvaluationsRequest(
+        user_app_id=self.user_app_id,
+        eval_metrics=[eval_metric],
+    )
+    response = self._grpc_request(self.STUB.PostEvaluations, request)
+    if response.status.code != status_code_pb2.SUCCESS:
+      raise Exception(response.status)
+    self.logger.info(
+        "\nModel evaluation in progress. Kindly allow a few minutes for completion. Processing time may vary based on the model and dataset sizes."
+    )
+
+    return response.eval_metrics
+
+  def get_eval_by_id(
+      self,
+      eval_id: str,
+      label_counts=False,
+      test_set=False,
+      binary_metrics=False,
+      confusion_matrix=False,
+      metrics_by_class=False,
+      metrics_by_area=False,
+  ) -> resources_pb2.EvalMetrics:
+    """Get detail eval_metrics by eval_id with extra metric fields
+
+    Args:
+        eval_id (str): eval id
+        label_counts (bool, optional): Set True to get label counts. Defaults to False.
+        test_set (bool, optional): Set True to get test set. Defaults to False.
+        binary_metrics (bool, optional): Set True to get binary metric. Defaults to False.
+        confusion_matrix (bool, optional): Set True to get confusion matrix. Defaults to False.
+        metrics_by_class (bool, optional): Set True to get metrics by class. Defaults to False.
+        metrics_by_area (bool, optional): Set True to get metrics by area. Defaults to False.
+
+    Raises:
+        Exception: Failed to call API
+
+    Returns:
+        resources_pb2.EvalMetrics: eval_metrics
+    """
+    request = service_pb2.GetEvaluationRequest(
+        user_app_id=self.user_app_id,
+        evaluation_id=eval_id,
+        fields=resources_pb2.FieldsValue(
+            label_counts=label_counts,
+            test_set=test_set,
+            binary_metrics=binary_metrics,
+            confusion_matrix=confusion_matrix,
+            metrics_by_class=metrics_by_class,
+            metrics_by_area=metrics_by_area,
+        ))
+    response = self._grpc_request(self.STUB.GetEvaluation, request)
+
+    if response.status.code != status_code_pb2.SUCCESS:
+      raise Exception(response.status)
+
+    return response.eval_metrics
+
+  def get_latest_eval(self,
+                      label_counts=False,
+                      test_set=False,
+                      binary_metrics=False,
+                      confusion_matrix=False,
+                      metrics_by_class=False,
+                      metrics_by_area=False) -> Union[resources_pb2.EvalMetrics, None]:
+    """
+    Run `get_eval_by_id` method with latest `eval_id`
+
+    Args:
+      label_counts (bool, optional): Set True to get label counts. Defaults to False.
+      test_set (bool, optional): Set True to get test set. Defaults to False.
+      binary_metrics (bool, optional): Set True to get binary metric. Defaults to False.
+      confusion_matrix (bool, optional): Set True to get confusion matrix. Defaults to False.
+      metrics_by_class (bool, optional): Set True to get metrics by class. Defaults to False.
+      metrics_by_area (bool, optional): Set True to get metrics by area. Defaults to False.
+
+    Returns:
+      eval_metric if model is evaluated otherwise None.
+
+    """
+
+    _latest = self.list_evaluations()[0]
+    result = None
+    if _latest.status.code == status_code_pb2.MODEL_EVALUATED:
+      result = self.get_eval_by_id(
+          eval_id=_latest.id,
+          label_counts=label_counts,
+          test_set=test_set,
+          binary_metrics=binary_metrics,
+          confusion_matrix=confusion_matrix,
+          metrics_by_class=metrics_by_class,
+          metrics_by_area=metrics_by_area)
+
+    return result

clarifai/client/module.py

@@ -18,6 +18,7 @@ class Module(Lister, BaseClient):
                module_version: Dict = {'id': ""},
                base_url: str = "https://api.clarifai.com",
                pat: str = None,
+               token: str = None,
                **kwargs):
     """Initializes a Module object.
 
@@ -26,7 +27,8 @@ class Module(Lister, BaseClient):
             module_id (str): The Module ID to interact with.
             module_version (dict): The Module Version to interact with.
             base_url (str): Base API url. Default "https://api.clarifai.com"
-            pat (str): A personal access token for authentication. Can be set as env var CLARIFAI_PAT
+            pat (str): A personal access token for authentication. Can be set as env var CLARIFAI_PAT.
+            token (str): A session token for authentication. Accepts either a session token or a pat. Can be set as env var CLARIFAI_SESSION_TOKEN.
             **kwargs: Additional keyword arguments to be passed to the Module.
         """
     if url and module_id:
@@ -40,8 +42,9 @@ class Module(Lister, BaseClient):
 
     self.kwargs = {**kwargs, 'id': module_id, 'module_version': module_version}
     self.module_info = resources_pb2.Module(**self.kwargs)
-    self.logger = get_logger(logger_level="INFO")
-    BaseClient.__init__(self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat)
+    self.logger = get_logger(logger_level="INFO", name=__name__)
+    BaseClient.__init__(
+        self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat, token=token)
     Lister.__init__(self)
 
   def list_versions(self, page_no: int = None,
@@ -78,10 +81,9 @@ class Module(Lister, BaseClient):
     for module_version_info in all_module_versions_info:
       module_version_info['id'] = module_version_info['module_version_id']
       del module_version_info['module_version_id']
-      yield Module(
+      yield Module.from_auth_helper(
+          self.auth_helper,
           module_id=self.id,
-          base_url=self.base,
-          pat=self.pat,
           **dict(self.kwargs, module_version=module_version_info))
 
   def __getattr__(self, name):

clarifai/client/runner.py

@@ -39,6 +39,7 @@ class Runner(BaseClient):
                check_runner_exists: bool = True,
                base_url: str = "https://api.clarifai.com",
                pat: str = None,
+               token: str = None,
                num_parallel_polls: int = 4,
                **kwargs) -> None:
     """
@@ -47,6 +48,7 @@ class Runner(BaseClient):
       user_id (str): Clarifai User ID
       base_url (str): Base API url. Default "https://api.clarifai.com"
       pat (str): A personal access token for authentication. Can be set as env var CLARIFAI_PAT
+      token (str): A session token for authentication. Accepts either a session token or a pat. Can be set as env var CLARIFAI_SESSION_TOKEN
       num_parallel_polls (int): the max number of threads for parallel run loops to be fetching work from
     """
     user_id = user_id or os.environ.get("CLARIFAI_USER_ID", "")
@@ -60,7 +62,7 @@ class Runner(BaseClient):
     self.kwargs = {**kwargs, 'id': runner_id, 'user_id': user_id}
     self.runner_info = resources_pb2.Runner(**self.kwargs)
     self.num_parallel_polls = min(10, num_parallel_polls)
-    BaseClient.__init__(self, user_id=self.user_id, app_id="", base=base_url, pat=pat)
+    BaseClient.__init__(self, user_id=self.user_id, app_id="", base=base_url, pat=pat, token=token)
 
     # Check that the runner exists.
     if check_runner_exists:

clarifai/client/search.py

@@ -23,7 +23,8 @@ class Search(Lister, BaseClient):
                top_k: int = DEFAULT_TOP_K,
                metric: str = DEFAULT_SEARCH_METRIC,
                base_url: str = "https://api.clarifai.com",
-               pat: str = None):
+               pat: str = None,
+               token: str = None):
     """Initialize the Search object.
 
     Args:
@@ -33,6 +34,7 @@ class Search(Lister, BaseClient):
         metric (str, optional): Similarity metric (either 'cosine' or 'euclidean'). Defaults to 'cosine'.
         base_url (str, optional): Base API url. Defaults to "https://api.clarifai.com".
         pat (str, optional): A personal access token for authentication. Can be set as env var CLARIFAI_PAT
+        token (str): A session token for authentication. Accepts either a session token or a pat. Can be set as env var CLARIFAI_SESSION_TOKEN
 
     Raises:
         UserError: If the metric is not 'cosine' or 'euclidean'.
@@ -46,9 +48,10 @@ class Search(Lister, BaseClient):
     self.data_proto = resources_pb2.Data()
     self.top_k = top_k
 
-    self.inputs = Inputs(user_id=self.user_id, app_id=self.app_id, pat=pat)
+    self.inputs = Inputs(user_id=self.user_id, app_id=self.app_id, pat=pat, token=token)
     self.rank_filter_schema = get_schema()
-    BaseClient.__init__(self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat)
+    BaseClient.__init__(
+        self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat, token=token)
     Lister.__init__(self, page_size=1000)
 
   def _get_annot_proto(self, **kwargs):

clarifai/client/user.py

@@ -19,6 +19,7 @@ class User(Lister, BaseClient):
                user_id: str = None,
                base_url: str = "https://api.clarifai.com",
                pat: str = None,
+               token: str = None,
                **kwargs):
     """Initializes an User object.
 
@@ -26,12 +27,13 @@ class User(Lister, BaseClient):
         user_id (str): The user ID for the user to interact with.
         base_url (str): Base API url. Default "https://api.clarifai.com"
         pat (str): A personal access token for authentication. Can be set as env var CLARIFAI_PAT
+        token (str): A session token for authentication. Accepts either a session token or a pat. Can be set as env var CLARIFAI_SESSION_TOKEN
         **kwargs: Additional keyword arguments to be passed to the User.
     """
     self.kwargs = {**kwargs, 'id': user_id}
     self.user_info = resources_pb2.User(**self.kwargs)
     self.logger = get_logger(logger_level="INFO", name=__name__)
-    BaseClient.__init__(self, user_id=self.id, app_id="", base=base_url, pat=pat)
+    BaseClient.__init__(self, user_id=self.id, app_id="", base=base_url, pat=pat, token=token)
     Lister.__init__(self)
 
   def list_apps(self, filter_by: Dict[str, Any] = {}, page_no: int = None,
@@ -62,7 +64,9 @@ class User(Lister, BaseClient):
         per_page=per_page,
         page_no=page_no)
     for app_info in all_apps_info:
-      yield App(base_url=self.base, pat=self.pat, **app_info)
+      yield App.from_auth_helper(
+          self.auth_helper,
+          **app_info)  #(base_url=self.base, pat=self.pat, token=self.token, **app_info)
 
   def list_runners(self, filter_by: Dict[str, Any] = {}, page_no: int = None,
                    per_page: int = None) -> Generator[Runner, None, None]:
@@ -94,7 +98,8 @@ class User(Lister, BaseClient):
         page_no=page_no)
 
     for runner_info in all_runners_info:
-      yield Runner(check_runner_exists=False, base_url=self.base, pat=self.pat, **runner_info)
+      yield Runner.from_auth_helper(
+          auth=self.auth_helper, check_runner_exists=False, **runner_info)
 
   def create_app(self, app_id: str, base_workflow: str = 'Empty', **kwargs) -> App:
     """Creates an app for the user.
@@ -120,8 +125,7 @@ class User(Lister, BaseClient):
     if response.status.code != status_code_pb2.SUCCESS:
       raise Exception(response.status)
     self.logger.info("\nApp created\n%s", response.status)
-    kwargs.update({'user_id': self.id, 'base_url': self.base, 'pat': self.pat})
-    return App(app_id=app_id, **kwargs)
+    return App.from_auth_helper(auth=self.auth_helper, app_id=app_id)
 
   def create_runner(self, runner_id: str, labels: List[str], description: str) -> Runner:
     """Create a runner
@@ -151,14 +155,13 @@ class User(Lister, BaseClient):
       raise Exception(response.status)
     self.logger.info("\nRunner created\n%s", response.status)
 
-    return Runner(
+    return Runner.from_auth_helper(
+        auth=self.auth_helper,
         runner_id=runner_id,
         user_id=self.id,
         labels=labels,
         description=description,
-        check_runner_exists=False,
-        base_url=self.base,
-        pat=self.pat)
+        check_runner_exists=False)
 
   def app(self, app_id: str, **kwargs) -> App:
     """Returns an App object for the specified app ID.
@@ -181,8 +184,7 @@ class User(Lister, BaseClient):
       raise Exception(response.status)
 
     kwargs['user_id'] = self.id
-    kwargs.update({'base_url': self.base, 'pat': self.pat})
-    return App(app_id=app_id, **kwargs)
+    return App.from_auth_helper(auth=self.auth_helper, app_id=app_id, **kwargs)
 
   def runner(self, runner_id: str) -> Runner:
     """Returns a Runner object if exists.
@@ -210,7 +212,7 @@ class User(Lister, BaseClient):
     kwargs = self.process_response_keys(dict_response[list(dict_response.keys())[1]],
                                         list(dict_response.keys())[1])
 
-    return Runner(check_runner_exists=False, base_url=self.base, pat=self.pat, **kwargs)
+    return Runner.from_auth_helper(self.auth_helper, check_runner_exists=False, **kwargs)
 
   def delete_app(self, app_id: str) -> None:
     """Deletes an app for the user.

clarifai/client/workflow.py

@@ -27,6 +27,7 @@ class Workflow(Lister, BaseClient):
                output_config: Dict = {'min_value': 0},
                base_url: str = "https://api.clarifai.com",
                pat: str = None,
+               token: str = None,
                **kwargs):
     """Initializes a Workflow object.
 
@@ -40,6 +41,8 @@ class Workflow(Lister, BaseClient):
           select_concepts (list[Concept]): The concepts to select.
           sample_ms (int): The number of milliseconds to sample.
         base_url (str): Base API url. Default "https://api.clarifai.com"
+        pat (str): A personal access token for authentication. Can be set as env var CLARIFAI_PAT
+        token (str): A session token for authentication. Accepts either a session token or a pat. Can be set as env var CLARIFAI_SESSION_TOKEN
         **kwargs: Additional keyword arguments to be passed to the Workflow.
     """
     if url and workflow_id:
@@ -54,8 +57,9 @@ class Workflow(Lister, BaseClient):
     self.kwargs = {**kwargs, 'id': workflow_id, 'version': workflow_version}
     self.output_config = output_config
     self.workflow_info = resources_pb2.Workflow(**self.kwargs)
-    self.logger = get_logger(logger_level="INFO")
-    BaseClient.__init__(self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat)
+    self.logger = get_logger(logger_level="INFO", name=__name__)
+    BaseClient.__init__(
+        self, user_id=self.user_id, app_id=self.app_id, base=base_url, pat=pat, token=token)
     Lister.__init__(self)
 
   def predict(self, inputs: List[Input], workflow_state_id: str = None):
@@ -206,10 +210,9 @@ class Workflow(Lister, BaseClient):
     for workflow_version_info in all_workflow_versions_info:
       workflow_version_info['id'] = workflow_version_info['workflow_version_id']
       del workflow_version_info['workflow_version_id']
-      yield Workflow(
+      yield Workflow.from_auth_helper(
+          auth=self.auth_helper,
           workflow_id=self.id,
-          base_url=self.base,
-          pat=self.pat,
           **dict(self.kwargs, version=workflow_version_info))
 
   def export(self, out_path: str):

clarifai/datasets/upload/features.py

@@ -20,6 +20,7 @@ class VisualClassificationFeatures:
   geo_info: Optional[List[float]] = None  #[Longitude, Latitude]
   id: Optional[int] = None  # image_id
   metadata: Optional[dict] = None
+  image_bytes: Optional[bytes] = None
 
 
 @dataclass
@@ -31,6 +32,7 @@ class VisualDetectionFeatures:
   geo_info: Optional[List[float]] = None  #[Longitude, Latitude]
   id: Optional[int] = None  # image_id
   metadata: Optional[dict] = None
+  image_bytes: Optional[bytes] = None
 
 
 @dataclass
@@ -42,3 +44,4 @@ class VisualSegmentationFeatures:
   geo_info: Optional[List[float]] = None  #[Longitude, Latitude]
   id: Optional[int] = None  # image_id
   metadata: Optional[dict] = None
+  image_bytes: Optional[bytes] = None

clarifai/datasets/upload/image.py

@@ -1,4 +1,5 @@
 import os
+import uuid
 from concurrent.futures import ThreadPoolExecutor
 from typing import List, Tuple, Type
 
@@ -31,22 +32,34 @@ class VisualClassificationDataset(ClarifaiDataset):
       image_path = data_item.image_path
       labels = data_item.labels if isinstance(data_item.labels,
                                               list) else [data_item.labels]  # clarifai concept
-      input_id = f"{self.dataset_id}-{id}" if data_item.id is None else f"{self.dataset_id}-{str(data_item.id)}"
+      input_id = f"{self.dataset_id}-{uuid.uuid4().hex[:8]}" if data_item.id is None else f"{self.dataset_id}-{str(data_item.id)}"
       geo_info = data_item.geo_info
       if data_item.metadata is not None:
         metadata.update(data_item.metadata)
-      else:
+      elif image_path is not None:
         metadata.update({"filename": os.path.basename(image_path)})
+      else:
+        metadata = None
 
       self.all_input_ids[id] = input_id
-      input_protos.append(
-          Inputs.get_input_from_file(
-              input_id=input_id,
-              image_file=image_path,
-              dataset_id=self.dataset_id,
-              labels=labels,
-              geo_info=geo_info,
-              metadata=metadata))
+      if data_item.image_bytes is not None:
+        input_protos.append(
+            Inputs.get_input_from_bytes(
+                input_id=input_id,
+                image_bytes=data_item.image_bytes,
+                dataset_id=self.dataset_id,
+                labels=labels,
+                geo_info=geo_info,
+                metadata=metadata))
+      else:
+        input_protos.append(
+            Inputs.get_input_from_file(
+                input_id=input_id,
+                image_file=image_path,
+                dataset_id=self.dataset_id,
+                labels=labels,
+                geo_info=geo_info,
+                metadata=metadata))
 
     with ThreadPoolExecutor(max_workers=4) as executor:
       futures = [executor.submit(process_data_item, id) for id in batch_input_ids]
@@ -79,7 +92,7 @@ class VisualDetectionDataset(ClarifaiDataset):
       image = data_item.image_path
       labels = data_item.labels  # list:[l1,...,ln]
       bboxes = data_item.bboxes  # [[xmin,ymin,xmax,ymax],...,[xmin,ymin,xmax,ymax]]
-      input_id = f"{self.dataset_id}-{id}" if data_item.id is None else f"{self.dataset_id}-{str(data_item.id)}"
+      input_id = f"{self.dataset_id}-{uuid.uuid4().hex[:8]}" if data_item.id is None else f"{self.dataset_id}-{str(data_item.id)}"
       if data_item.metadata is not None:
         metadata.update(data_item.metadata)
       else:
@@ -87,13 +100,22 @@ class VisualDetectionDataset(ClarifaiDataset):
       geo_info = data_item.geo_info
 
       self.all_input_ids[id] = input_id
-      input_protos.append(
-          Inputs.get_input_from_file(
-              input_id=input_id,
-              image_file=image,
-              dataset_id=self.dataset_id,
-              geo_info=geo_info,
-              metadata=metadata))
+      if data_item.image_bytes is not None:
+        input_protos.append(
+            Inputs.get_input_from_bytes(
+                input_id=input_id,
+                image_bytes=data_item.image_bytes,
+                dataset_id=self.dataset_id,
+                geo_info=geo_info,
+                metadata=metadata))
+      else:
+        input_protos.append(
+            Inputs.get_input_from_file(
+                input_id=input_id,
+                image_file=image,
+                dataset_id=self.dataset_id,
+                geo_info=geo_info,
+                metadata=metadata))
       # iter over bboxes and labels
       # one id could have more than one bbox and label
       for i in range(len(bboxes)):
@@ -131,7 +153,7 @@ class VisualSegmentationDataset(ClarifaiDataset):
       image = data_item.image_path
       labels = data_item.labels
       _polygons = data_item.polygons  # list of polygons: [[[x,y],...,[x,y]],...]
-      input_id = f"{self.dataset_id}-{id}" if data_item.id is None else f"{self.dataset_id}-{str(data_item.id)}"
+      input_id = f"{self.dataset_id}-{uuid.uuid4().hex[:8]}" if data_item.id is None else f"{self.dataset_id}-{str(data_item.id)}"
       if data_item.metadata is not None:
         metadata.update(data_item.metadata)
       else:
@@ -139,13 +161,22 @@ class VisualSegmentationDataset(ClarifaiDataset):
       geo_info = data_item.geo_info
 
       self.all_input_ids[id] = input_id
-      input_protos.append(
-          Inputs.get_input_from_file(
-              input_id=input_id,
-              image_file=image,
-              dataset_id=self.dataset_id,
-              geo_info=geo_info,
-              metadata=metadata))
+      if data_item.image_bytes is not None:
+        input_protos.append(
+            Inputs.get_input_from_bytes(
+                input_id=input_id,
+                image_bytes=data_item.image_bytes,
+                dataset_id=self.dataset_id,
+                geo_info=geo_info,
+                metadata=metadata))
+      else:
+        input_protos.append(
+            Inputs.get_input_from_file(
+                input_id=input_id,
+                image_file=image,
+                dataset_id=self.dataset_id,
+                geo_info=geo_info,
+                metadata=metadata))
 
       ## Iterate over each masked image and create a proto for upload to clarifai
       ## The length of masks/polygons-list and labels must be equal

clarifai/datasets/upload/loaders/README.md

@@ -8,15 +8,15 @@ If a dataset module exists in the zoo, uploading the specific dataset can be eas
 
 ```python
 from clarifai.client.app import App
-from clarifai.datasets.upload.loaders.coco_segmentation import COCOSegmentationDataLoader
+from clarifai.datasets.upload.loaders.coco_detection import COCODetectionDataLoader
 
 app = App(app_id="", user_id="")
 # Create a dataset in Clarifai App
 dataset = app.create_dataset(dataset_id="")
 # instantiate dataloader object
-coco_seg_dataloader = COCOSegmentationDataLoader(images_dir="", label_filepath="")
+coco_det_dataloader = COCODetectionDataLoader(images_dir="", label_filepath="")
 # execute data upload to Clarifai app dataset
-dataset.upload_dataset(dataloader=coco_seg_dataloader)
+dataset.upload_dataset(dataloader=coco_det_dataloader)
 ```
 
 ## Dataset Loaders
@@ -24,7 +24,6 @@ dataset.upload_dataset(dataloader=coco_seg_dataloader)
  | dataset name | task | module name (.py)
  | --- | --- | ---
  | [COCO 2017](https://cocodataset.org/#download) | Detection | `coco_detection` |
- |        | Segmentation | `coco_segmentation` |
  |       | Captions | `coco_captions` |
  |[xVIEW](http://xviewdataset.org/)  | Detection | `xview_detection` |
  | [ImageNet](https://www.image-net.org/)  | Classification | `imagenet_classification` |