orca-sdk 0.0.98__py3-none-any.whl → 0.0.100__py3-none-any.whl

orca_sdk/client.py

@@ -263,8 +263,8 @@ class MemoryMetrics(TypedDict):
     neighbor_predicted_label_matches_current_label: NotRequired[bool | None]
     spread: NotRequired[float]
     uniformity: NotRequired[float]
-    subconcept_cluster_id: NotRequired[int | None]
-    subconcept_name: NotRequired[str | None]
+    concept_id: NotRequired[int | None]
+    subconcept_id: NotRequired[int | None]
 
 
 MemoryType = Literal["LABELED", "SCORED"]
@@ -301,6 +301,19 @@ class MemorysetClusterMetrics(TypedDict):
     num_clusters: int
 
 
+class MemorysetConceptAnalysisConfig(TypedDict):
+    high_level_description: NotRequired[str | None]
+    max_sample_rows: NotRequired[int]
+    max_trial_count: NotRequired[int]
+    min_desired_clusters_per_label: NotRequired[int]
+    max_desired_clusters_per_label: NotRequired[int]
+    accuracy_importance: NotRequired[float]
+    noise_penalty: NotRequired[float]
+    naming_examples_count: NotRequired[int]
+    naming_counterexample_count: NotRequired[int]
+    seed: NotRequired[int]
+
+
 class MemorysetDuplicateAnalysisConfig(TypedDict):
     potential_duplicate_threshold: NotRequired[float]
 
@@ -339,19 +352,6 @@ class MemorysetProjectionMetrics(TypedDict):
     pass
 
 
-class MemorysetSubconceptAnalysisConfig(TypedDict):
-    high_level_description: NotRequired[str | None]
-    max_sample_rows: NotRequired[int]
-    max_trial_count: NotRequired[int]
-    min_desired_clusters_per_label: NotRequired[int]
-    max_desired_clusters_per_label: NotRequired[int]
-    accuracy_importance: NotRequired[float]
-    noise_penalty: NotRequired[float]
-    naming_examples_count: NotRequired[int]
-    naming_counterexample_count: NotRequired[int]
-    seed: NotRequired[int]
-
-
 class MemorysetUpdate(TypedDict):
     label_names: NotRequired[list[str]]
     description: NotRequired[str | None]
@@ -628,11 +628,11 @@ class ServiceUnavailableErrorResponse(TypedDict):
     service: str
 
 
-class SubconceptClusterMetrics(TypedDict):
-    cluster_id: int
+class SubConceptMetrics(TypedDict):
+    id: int
     name: str
-    primary_label_index: NotRequired[int | None]
-    description: NotRequired[str | None]
+    description: str | None
+    primary_label: int | None
     memory_count: int
 
 
@@ -840,6 +840,13 @@ class GetDatasourceByNameOrIdEmbeddingEvaluationByTaskIdParams(TypedDict):
 
 class GetDatasourceByNameOrIdDownloadParams(TypedDict):
     name_or_id: str
+    file_type: NotRequired[Literal["hf_dataset", "json", "csv"]]
+    """
+    File type to download: 
+    * `hf_dataset`: Zipped HuggingFace dataset (default)
+    * `json`: Row-oriented JSON array
+    * `csv`: CSV file
+    """
 
 
 class PatchClassificationModelByNameOrIdParams(TypedDict):
@@ -938,7 +945,6 @@ class GetTelemetryPredictionByPredictionIdExplanationParams(TypedDict):
 
 class GetTelemetryPredictionByPredictionIdActionParams(TypedDict):
     prediction_id: str
-    refresh: NotRequired[bool]
 
 
 class GetTelemetryPredictionByPredictionIdMemorySuggestionsParams(TypedDict):
@@ -1100,6 +1106,15 @@ class ColumnInfo(TypedDict):
     int_values: NotRequired[list[int] | None]
 
 
+class ConceptMetrics(TypedDict):
+    id: int
+    name: str
+    description: str | None
+    primary_label: int | None
+    memory_count: int
+    subconcepts: list[SubConceptMetrics]
+
+
 class CreateClassificationModelRequest(TypedDict):
     name: str
     description: NotRequired[str | None]
@@ -1378,7 +1393,7 @@ class MemorysetAnalysisConfigs(TypedDict):
     projection: NotRequired[MemorysetProjectionAnalysisConfig | None]
     cluster: NotRequired[MemorysetClusterAnalysisConfig | None]
     class_patterns: NotRequired[MemorysetClassPatternsAnalysisConfig | None]
-    subconcepts: NotRequired[MemorysetSubconceptAnalysisConfig | None]
+    concepts: NotRequired[MemorysetConceptAnalysisConfig | None]
 
 
 class MemorysetAnalysisRequest(TypedDict):
@@ -1388,12 +1403,21 @@ class MemorysetAnalysisRequest(TypedDict):
     configs: MemorysetAnalysisConfigs
 
 
-class MemorysetSubconceptMetrics(TypedDict):
-    clusters_by_id: dict[str, SubconceptClusterMetrics]
-    num_clusters: int
+class MemorysetConceptMetrics(TypedDict):
+    concepts: list[ConceptMetrics]
     num_outliers: int
 
 
+class MemorysetMetrics(TypedDict):
+    neighbor: NotRequired[MemorysetNeighborMetrics | None]
+    label: NotRequired[MemorysetLabelMetrics | None]
+    duplicate: NotRequired[MemorysetDuplicateMetrics | None]
+    projection: NotRequired[MemorysetProjectionMetrics | None]
+    cluster: NotRequired[MemorysetClusterMetrics | None]
+    class_patterns: NotRequired[MemorysetClassPatternsMetrics | None]
+    concepts: NotRequired[MemorysetConceptMetrics | None]
+
+
 class PaginatedUnionLabeledMemoryWithFeedbackMetricsScoredMemoryWithFeedbackMetrics(TypedDict):
     items: list[LabeledMemoryWithFeedbackMetrics | ScoredMemoryWithFeedbackMetrics]
     total: int
@@ -1449,34 +1473,6 @@ class EmbeddingEvaluationResult(TypedDict):
     evaluation_results: list[EmbeddingModelResult]
 
 
-class MemorysetMetrics(TypedDict):
-    neighbor: NotRequired[MemorysetNeighborMetrics | None]
-    label: NotRequired[MemorysetLabelMetrics | None]
-    duplicate: NotRequired[MemorysetDuplicateMetrics | None]
-    projection: NotRequired[MemorysetProjectionMetrics | None]
-    cluster: NotRequired[MemorysetClusterMetrics | None]
-    class_patterns: NotRequired[MemorysetClassPatternsMetrics | None]
-    subconcepts: NotRequired[MemorysetSubconceptMetrics | None]
-
-
-class PaginatedTask(TypedDict):
-    items: list[Task]
-    total: int
-    offset: int
-    limit: int
-
-
-class DatasourceEmbeddingEvaluationsResponse(TypedDict):
-    task_id: str
-    org_id: str
-    datasource_id: str
-    status: TaskStatus
-    result: EmbeddingEvaluationResult | None
-    payload: DatasourceEmbeddingEvaluationsTaskPayload
-    created_at: str
-    updated_at: str
-
-
 class MemorysetAnalysisResponse(TypedDict):
     task_id: str
     org_id: str
@@ -1516,6 +1512,13 @@ class MemorysetMetadata(TypedDict):
     query_prompt_override: str | None
 
 
+class PaginatedTask(TypedDict):
+    items: list[Task]
+    total: int
+    offset: int
+    limit: int
+
+
 class BootstrapClassificationModelMeta(TypedDict):
     datasource_meta: DatasourceMetadata
     memoryset_meta: MemorysetMetadata
@@ -1531,6 +1534,17 @@ class BootstrapClassificationModelResponse(TypedDict):
     input: BootstrapClassificationModelRequest | None
 
 
+class DatasourceEmbeddingEvaluationsResponse(TypedDict):
+    task_id: str
+    org_id: str
+    datasource_id: str
+    status: TaskStatus
+    result: EmbeddingEvaluationResult | None
+    payload: DatasourceEmbeddingEvaluationsTaskPayload
+    created_at: str
+    updated_at: str
+
+
 class OrcaClient(Client):
     @staticmethod
     def _parse_params(
@@ -1911,6 +1925,40 @@ class OrcaClient(Client):
         """Get an embedding evaluation task by ID."""
         pass
 
+    @overload
+    def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}/download"],
+        *,
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: Literal["json"] = "json",
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> list[dict[str, Any]]:
+        """Download datasource in the specified format."""
+        pass
+
+    @overload
+    def GET(
+        self,
+        path: Literal["/datasource/{name_or_id}/download"],
+        *,
+        params: GetDatasourceByNameOrIdDownloadParams,
+        parse_as: Literal["text"],
+        headers: HeaderTypes | None = None,
+        cookies: CookieTypes | None = None,
+        auth: AuthTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        follow_redirects: bool | UseClientDefault = USE_CLIENT_DEFAULT,
+        timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
+        extensions: RequestExtensions | None = None,
+    ) -> str:
+        """Download datasource in the specified format."""
+        pass
+
     @overload
     def GET(
         self,
@@ -1925,7 +1973,7 @@ class OrcaClient(Client):
         timeout: TimeoutTypes | UseClientDefault = USE_CLIENT_DEFAULT,
         extensions: RequestExtensions | None = None,
     ) -> bytes:
-        """Streams a zipped dataset file response"""
+        """Download datasource in the specified format."""
         pass
 
     @overload

orca_sdk/datasource.py

@@ -1,6 +1,7 @@
 from __future__ import annotations
 
 import logging
+import os
 import tempfile
 import zipfile
 from datetime import datetime
@@ -141,28 +142,6 @@ class Datasource:
             + "})"
         )
 
-    def download(self, output_path: str | PathLike) -> None:
-        """
-        Download the datasource as a ZIP and extract them to a specified path.
-
-        Params:
-            output_path: The local file path or directory where the downloaded files will be saved.
-
-        Returns:
-            None
-        """
-        # TODO: add progress bar to the download
-        response = orca_api.GET("/datasource/{name_or_id}/download", params={"name_or_id": self.id}, parse_as=None)
-        with tempfile.NamedTemporaryFile(suffix=".zip") as tmp_zip:
-            tmp_zip.write(response)
-            tmp_zip.flush()
-            with zipfile.ZipFile(tmp_zip.name, "r") as zf:
-                Path(output_path).mkdir(parents=True, exist_ok=True)
-                for file in zf.namelist():
-                    out_file = Path(output_path) / Path(file).name
-                    with zf.open(file) as af:
-                        out_file.write_bytes(af.read())
-
     @classmethod
     def from_hf_dataset(
         cls, name: str, dataset: Dataset, if_exists: CreateMode = "error", description: str | None = None
@@ -501,3 +480,45 @@ class Datasource:
 
     def __len__(self) -> int:
         return self.length
+
+    def download(
+        self, output_dir: str | PathLike, file_type: Literal["hf_dataset", "json", "csv"] = "hf_dataset"
+    ) -> None:
+        """
+        Download the datasource to a specified path in the specified format type
+
+        Params:
+            output_dir: The local directory where the downloaded file will be saved.
+            file_type: The type of file to download.
+
+        Returns:
+            None
+        """
+        extension = "zip" if file_type == "hf_dataset" else file_type
+        output_path = Path(output_dir) / f"{self.name}.{extension}"
+        with open(output_path, "wb") as download_file:
+            with orca_api.stream("GET", f"/datasource/{self.id}/download", params={"file_type": file_type}) as response:
+                total_chunks = int(response.headers["X-Total-Chunks"]) if "X-Total-Chunks" in response.headers else None
+                with tqdm(desc=f"Downloading", total=total_chunks, disable=total_chunks is None) as progress:
+                    for chunk in response.iter_bytes():
+                        download_file.write(chunk)
+                        progress.update(1)
+
+        # extract the zip file
+        if extension == "zip":
+            extract_dir = Path(output_dir) / self.name
+            with zipfile.ZipFile(output_path, "r") as zip_ref:
+                zip_ref.extractall(extract_dir)
+            output_path.unlink()  # Remove the zip file after extraction
+            logging.info(f"Downloaded {extract_dir}")
+        else:
+            logging.info(f"Downloaded {output_path}")
+
+    def to_list(self) -> list[dict]:
+        """
+        Convert the datasource to a list of dictionaries.
+
+        Returns:
+            A list of dictionaries representation of the datasource.
+        """
+        return orca_api.GET("/datasource/{name_or_id}/download", params={"name_or_id": self.id, "file_type": "json"})

orca_sdk/datasource_test.py

@@ -1,10 +1,14 @@
+import json
 import os
 import tempfile
+from typing import cast
 from uuid import uuid4
 
+import numpy as np
 import pandas as pd
 import pyarrow as pa
 import pytest
+from datasets import Dataset
 
 from .datasource import Datasource
 
@@ -99,13 +103,6 @@ def test_drop_datasource_invalid_input():
         Datasource.drop("not valid id")
 
 
-def test_download_datasource(datasource):
-    with tempfile.TemporaryDirectory() as temp_dir:
-        output_path = os.path.join(temp_dir, "datasource.zip")
-        datasource.download(output_path)
-        assert os.path.exists(output_path)
-
-
 def test_from_list():
     # Test creating datasource from list of dictionaries
     data = [
@@ -296,3 +293,40 @@ def test_from_disk_already_exists():
             assert datasource2.name == datasource1.name
         finally:
             os.unlink(f.name)
+
+
+def test_to_list(hf_dataset, datasource):
+    assert datasource.to_list() == hf_dataset.to_list()
+
+
+def test_download_datasource(hf_dataset, datasource):
+    with tempfile.TemporaryDirectory() as temp_dir:
+        # Dataset download
+        datasource.download(temp_dir)
+        downloaded_hf_dataset_dir = f"{temp_dir}/{datasource.name}"
+        assert os.path.exists(downloaded_hf_dataset_dir)
+        assert os.path.isdir(downloaded_hf_dataset_dir)
+        assert not os.path.exists(f"{downloaded_hf_dataset_dir}.zip")
+        dataset_from_downloaded_hf_dataset = Dataset.load_from_disk(downloaded_hf_dataset_dir)
+        assert dataset_from_downloaded_hf_dataset.column_names == hf_dataset.column_names
+        assert dataset_from_downloaded_hf_dataset.to_dict() == hf_dataset.to_dict()
+
+        # JSON download
+        datasource.download(temp_dir, file_type="json")
+        downloaded_json_file = f"{temp_dir}/{datasource.name}.json"
+        assert os.path.exists(downloaded_json_file)
+        with open(downloaded_json_file, "r") as f:
+            content = json.load(f)
+            assert content == hf_dataset.to_list()
+
+        # CSV download
+        datasource.download(temp_dir, file_type="csv")
+        downloaded_csv_file = f"{temp_dir}/{datasource.name}.csv"
+        assert os.path.exists(downloaded_csv_file)
+        dataset_from_downloaded_csv = cast(Dataset, Dataset.from_csv(downloaded_csv_file))
+        assert dataset_from_downloaded_csv.column_names == hf_dataset.column_names
+        assert (
+            dataset_from_downloaded_csv.remove_columns("score").to_dict()
+            == hf_dataset.remove_columns("score").to_dict()
+        )
+        assert np.allclose(dataset_from_downloaded_csv["score"], hf_dataset["score"])

orca_sdk/telemetry.py

@@ -562,7 +562,7 @@ class ClassificationPrediction(_Prediction):
 
         response = orca_api.GET(
             "/telemetry/prediction/{prediction_id}/action",
-            params={"prediction_id": self.prediction_id, "refresh": refresh},
+            params={"prediction_id": self.prediction_id},
             timeout=30,
         )
         return (response["action"], response["rationale"])

{orca_sdk-0.0.98.dist-info → orca_sdk-0.0.100.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: orca_sdk
-Version: 0.0.98
+Version: 0.0.100
 Summary: SDK for interacting with Orca Services
 License: Apache-2.0
 Author: Orca DB Inc.

{orca_sdk-0.0.98.dist-info → orca_sdk-0.0.100.dist-info}/RECORD

@@ -19,12 +19,12 @@ orca_sdk/_utils/value_parser.py,sha256=c3qMABCCDQcIjn9N1orYYnlRwDW9JWdGwW_2TDZPL
 orca_sdk/_utils/value_parser_test.py,sha256=OybsiC-Obi32RRi9NIuwrVBRAnlyPMV1xVAaevSrb7M,1079
 orca_sdk/classification_model.py,sha256=SUiUgv_o3UUngpz3Le_L6DsijhjXVEB3yo84hrD1MX4,31172
 orca_sdk/classification_model_test.py,sha256=WganVoP-0vw1cqiVWJ2vXyGi4lwYp_hbZHultpxvFqk,19536
-orca_sdk/client.py,sha256=LEqdqMHbidh9NeZ_42RbEDR64KqXnqnX0JBjaoo2pnY,115704
+orca_sdk/client.py,sha256=gZ0e0gHoAxpvtsGYRnqH80XAZzi1EaYy2A0BplouOBA,117268
 orca_sdk/conftest.py,sha256=LHA46gDU_D0T_ogS6XOVQvGDDMD01nVZFWVcBYDConc,8885
 orca_sdk/credentials.py,sha256=KrmgP_5uqBKJXFJV6utTHIhU2odsr95VEqEXWe277DY,5074
 orca_sdk/credentials_test.py,sha256=whUweSJIEws6C8Bku-5V31Dv9hD1mFDDW7X2cCsB6g0,1629
-orca_sdk/datasource.py,sha256=344gqZsSV_N3RDlqXuLnDrPSizvciaAsSPmsiljNSmI,19329
-orca_sdk/datasource_test.py,sha256=nKQGjhX0VwCCLCdwl1ns-6kA5Ow-8pWQkSS9WV3CVww,9975
+orca_sdk/datasource.py,sha256=6wARRq-eNDJVSBABdVzn41z7s69xasTsqbozaVAsf9U,20263
+orca_sdk/datasource_test.py,sha256=wENPourrJvQN-uJJPaJI9EDuof6wVw3GirOhbrY4sFI,11564
 orca_sdk/embedding_model.py,sha256=YxMXdZ3tvnxnK93nArOr_HZ6QoRB-Mc5VNQJ0mcIdpk,26021
 orca_sdk/embedding_model_test.py,sha256=1aELyCuIzxSxUg7Z4tYtNOd8-hV5hFb-gsZTNh712OQ,7765
 orca_sdk/job.py,sha256=wWJPkkQbkNu_ylBtZN4AscU00VwWTfqlSmysRBUlivw,12787
@@ -33,8 +33,8 @@ orca_sdk/memoryset.py,sha256=hnhgHxsuV5JEDMnxazmO03VjxzPBENtHDPppHRfh9z4,85354
 orca_sdk/memoryset_test.py,sha256=14WG6u_adVmm6CSn2dM5lEfQYxybwh3s9Q7RBeTuoPE,20486
 orca_sdk/regression_model.py,sha256=je2g1BmoPCuouv5iWqDolja90F2w2vD6TooXA8KjL7c,24552
 orca_sdk/regression_model_test.py,sha256=8LDhtQeh52grZQ2Xd0ju1MQvb_hwosY_ORDDE3wS2LA,14570
-orca_sdk/telemetry.py,sha256=jOOFYYr1s3i5EASbCcmUJ_O469xCK5OFg7NVcVfTAlU,25400
+orca_sdk/telemetry.py,sha256=dRyf8fIvThkSBLDyD8BYkixg3nphsN-HbneWq7nbp_4,25380
 orca_sdk/telemetry_test.py,sha256=eT66C5lFdNg-pQdo2I__BP7Tn5fTc9aTkVo9ZhWwhU0,5519
-orca_sdk-0.0.98.dist-info/METADATA,sha256=z5MKWnC226a5mzHHN7aKJFvpFm9qTe46-gtaRTH8ppM,3613
-orca_sdk-0.0.98.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
-orca_sdk-0.0.98.dist-info/RECORD,,
+orca_sdk-0.0.100.dist-info/METADATA,sha256=-PwEi3sZwlqJerxyIo_4Fu2tG548bxONpXXHr6USW4E,3614
+orca_sdk-0.0.100.dist-info/WHEEL,sha256=sP946D7jFCHeNz5Iq4fL4Lu-PrWrFsgfLXbbkciIZwg,88
+orca_sdk-0.0.100.dist-info/RECORD,,