reorganizing plus small docs updates (#123)

* reorganizing plus small docs updates

* fix unintended default api base url change

cleanlab_studio/internal/upload_helpers.py

@@ -10,7 +10,40 @@
 
 from .api import api
 from .dataset_source import DatasetSource
-from .types import JSONDict
+from .types import FieldSchemaDict, JSONDict
+
+
+def upload_dataset(
+    api_key: str,
+    dataset_source: DatasetSource,
+    *,
+    schema_overrides: Optional[FieldSchemaDict] = None,
+    modality: Optional[str] = None,
+    id_column: Optional[str] = None,
+) -> str:
+    upload_id = upload_dataset_file(api_key, dataset_source)
+    schema = get_proposed_schema(api_key, upload_id)
+
+    if (schema is None or schema.get("immutable", False)) and (
+        schema_overrides is not None or modality is not None or id_column is not None
+    ):
+        raise ValueError(
+            "Schema_overrides, modality, and id_column parameters cannot be provided for simple zip uploads"
+        )
+
+    if schema is not None and not schema.get("immutable", False):
+        schema["metadata"]["name"] = dataset_source.dataset_name
+        if schema_overrides is not None:
+            for field in schema_overrides:
+                schema["fields"][field] = schema_overrides[field]
+        if modality is not None:
+            schema["metadata"]["modality"] = modality
+        if id_column is not None:
+            schema["metadata"]["id_column"] = id_column
+
+    api.confirm_schema(api_key, schema, upload_id)
+    dataset_id = get_ingestion_result(api_key, upload_id)
+    return dataset_id
 
 
 async def _upload_file_chunk_async(
@@ -58,7 +91,10 @@ def upload_file_parts(
 
 def upload_dataset_file(api_key: str, dataset_source: DatasetSource) -> str:
     upload_id, part_sizes, presigned_posts = api.initialize_upload(
-        api_key, dataset_source.get_filename(), dataset_source.file_type, dataset_source.file_size
+        api_key,
+        dataset_source.get_filename(),
+        dataset_source.file_type,
+        dataset_source.file_size,
     )
     upload_parts = upload_file_parts(dataset_source, part_sizes, presigned_posts)
     api.complete_file_upload(api_key, upload_id, upload_parts)

cleanlab_studio/studio/studio.py

@@ -7,7 +7,8 @@
 import numpy.typing as npt
 import pandas as pd
 
-from . import clean, upload, inference
+from . import inference
+from cleanlab_studio.internal import clean_helpers, upload_helpers
 from cleanlab_studio.internal.api import api
 from cleanlab_studio.internal.util import (
     init_dataset_source,
@@ -68,7 +69,7 @@ def upload_dataset(
             ID of uploaded dataset.
         """
         ds = init_dataset_source(dataset, dataset_name)
-        return upload.upload_dataset(
+        return upload_helpers.upload_dataset(
             self._api_key,
             ds,
             schema_overrides=schema_overrides,
@@ -266,7 +267,7 @@ def poll_cleanset_status(self, cleanset_id: str, timeout: Optional[int] = None)
         Returns:
             After cleanset is done being generated, returns `True` if cleanset is ready to use, `False` otherwise.
         """
-        return clean.poll_cleanset_status(self._api_key, cleanset_id, timeout)
+        return clean_helpers.poll_cleanset_status(self._api_key, cleanset_id, timeout)
 
     def get_latest_cleanset_id(self, project_id: str) -> str:
         """
@@ -298,7 +299,7 @@ def get_model(self, model_id: str) -> inference.Model:
             model_id: ID of model to get. This ID should be fetched in the deployments page of the app UI.
 
         Returns:
-            Model object with methods to run predictions on new input data
+            [Model](inference#class-model) object with methods to run predictions on new input data.
         """
         return inference.Model(self._api_key, model_id)