sinapsis-data-readers 0.1.17__tar.gz → 0.1.20__tar.gz

{sinapsis_data_readers-0.1.17/src/sinapsis_data_readers.egg-info → sinapsis_data_readers-0.1.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sinapsis-data-readers
-Version: 0.1.17
+Version: 0.1.20
 Summary: Templates to read data in different formats
 Author-email: SinapsisAI <dev@sinapsis.tech>
 Project-URL: Homepage, https://sinapsis.tech
@@ -14,8 +14,8 @@ Requires-Dist: sinapsis>=0.1.1
 Provides-Extra: nvidia-dali
 Requires-Dist: nvidia-dali-cuda120>=1.43.0; extra == "nvidia-dali"
 Provides-Extra: torch-codec
-Requires-Dist: torch<=2.5.1; extra == "torch-codec"
-Requires-Dist: torchcodec>=0.3.0; extra == "torch-codec"
+Requires-Dist: torch>=2.4.1; extra == "torch-codec"
+Requires-Dist: torchcodec>=0.0.3; extra == "torch-codec"
 Provides-Extra: sklearn-datasets
 Requires-Dist: pandas>=2.2.3; extra == "sklearn-datasets"
 Requires-Dist: scikit-learn>=1.5.2; extra == "sklearn-datasets"

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "sinapsis-data-readers"
-version = "0.1.17"
+version = "0.1.20"
 description = "Templates to read data in different formats"
 authors = [{ name = "SinapsisAI", email = "dev@sinapsis.tech" }]
 
@@ -14,8 +14,8 @@ dependencies = [
 [project.optional-dependencies]
 nvidia-dali = ["nvidia-dali-cuda120>=1.43.0"]
 torch-codec = [
-    "torch<=2.5.1",
-    "torchcodec>=0.3.0",
+    "torch>=2.4.1",
+    "torchcodec>=0.0.3",
 ]
 sklearn-datasets = ["pandas>=2.2.3", "scikit-learn>=1.5.2"]
 sktime-datareaders = ["sktime>=0.34.0"]

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20}/src/sinapsis_data_readers/templates/datasets_readers/dataset_splitter.py

@@ -38,7 +38,11 @@ class ImageDatasetSplit(BaseModel):
         """allow arbitrary types"""
 
         arbitrary_types_allowed = True
-
+        json_encoders : dict = {
+            pd.DataFrame: lambda df: df.to_dict(orient="records"),
+            pd.Series: lambda s: s.to_list(),
+            np.ndarray: lambda arr: arr.tolist(),
+        }
 
 class TabularDatasetSplit(BaseModel):
     """BaseModel to store the content of the data packets as a list

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20}/src/sinapsis_data_readers/templates/datasets_readers/sklearn_datasets.py

@@ -144,8 +144,7 @@ class SKLearnDatasets(BaseDynamicWrapperTemplate):
             X = results.iloc[:, :n_features]
             y = results.iloc[:, n_features:]
 
-        # x_vals = results.drop(columns=[TARGET], axis=1)
-        # y_vals = results[TARGET]
+
         x_train, x_test, y_train, y_test = train_test_split(X, y, train_size=split_size, random_state=0)
         split_data = TabularDatasetSplit(
             x_train=pd.DataFrame(x_train),
@@ -154,7 +153,7 @@ class SKLearnDatasets(BaseDynamicWrapperTemplate):
             y_test=pd.DataFrame(y_test),
         )
 
-        return split_data.model_dump()
+        return split_data.model_dump_json(indent=2)
 
     def execute(self, container: DataContainer) -> DataContainer:
         sklearn_dataset = self.wrapped_callable.__func__(**self.dataset_attributes.model_dump())

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20}/src/sinapsis_data_readers/templates/datasets_readers/sktime_datasets.py

@@ -104,13 +104,13 @@ class SKTimeDatasets(BaseDynamicWrapperTemplate):
             TabularDatasetSplit: Object containing the split time series data
         """
         y_train, y_test = temporal_train_test_split(dataset, train_size=self.attributes.train_size)
-        split_data =  TabularDatasetSplit(
+        split_dataset =  TabularDatasetSplit(
             x_train=pd.DataFrame(index=y_train.index),
             x_test=pd.DataFrame(index=y_test.index),
             y_train=pd.DataFrame(y_train),
             y_test=pd.DataFrame(y_test),
         )
-        return split_data.model_dump()
+        return split_dataset.model_dump_json(indent=2)
 
     def split_classification_dataset(self, X: Any, y: Any) -> TabularDatasetSplit:
         """Split a classification dataset into training and testing sets
@@ -126,15 +126,17 @@ class SKTimeDatasets(BaseDynamicWrapperTemplate):
             X_train, X_test, y_train, y_test = train_test_split(
                 X, y, train_size=self.attributes.train_size, random_state=0
             )
-            return TabularDatasetSplit(
+            split_dataset = TabularDatasetSplit(
                 x_train=pd.DataFrame(X_train),
                 x_test=pd.DataFrame(X_test),
                 y_train=pd.DataFrame(y_train),
                 y_test=pd.DataFrame(y_test),
             )
+            return split_dataset.model_dump_json(indent=2)
         except ValueError:
             self.logger.debug("Wrong format for split. original values")
-            return TabularDatasetSplit(x_train=pd.DataFrame(X), y_train=pd.DataFrame(y))
+            split_dataset = TabularDatasetSplit(x_train=pd.DataFrame(X), y_train=pd.DataFrame(y))
+            return split_dataset.model_dump_json(indent=2)
 
     def create_dataset(self):
         return self.wrapped_callable.__func__(**self.dataset_attributes.model_dump())

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20}/src/sinapsis_data_readers/templates/video_readers/video_reader_dali.py

@@ -134,19 +134,19 @@ class VideoReaderDali(BaseVideoReader):
             del self.video_reader
 
     def _read_video_frames(self) -> list[ImagePacket]:
-        """Reads video frames from the dali pipeline.
-
-        This method runs the video reader pipeline and adds the frames to a list
-        of ImagePacket objects.
-
-        Returns:
-            list[ImagePacket]: A list of ImagePacket objects representing the video frames.
-        """
+        """Reads video frames from the dali pipeline."""
         video_frames: list[ImagePacket] = []
         sequences_out = self.video_reader.run()
+        tensor_batch = sequences_out[0]
+
+        shape_result = tensor_batch.shape()
+        batch_size = shape_result[0][0]
 
-        for idx, frame in enumerate(sequences_out[0]):
+        for idx in range(batch_size):
+            frame_tensor = tensor_batch.at(idx)
+            frame = torch.as_tensor(frame_tensor, device="cuda")
             video_frames.append(self._make_image_packet(frame, frame_index=self.frame_count + idx))
+
         return video_frames
 
     def reset_state(self, template_name: str | None = None) -> None:

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20/src/sinapsis_data_readers.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sinapsis-data-readers
-Version: 0.1.17
+Version: 0.1.20
 Summary: Templates to read data in different formats
 Author-email: SinapsisAI <dev@sinapsis.tech>
 Project-URL: Homepage, https://sinapsis.tech
@@ -14,8 +14,8 @@ Requires-Dist: sinapsis>=0.1.1
 Provides-Extra: nvidia-dali
 Requires-Dist: nvidia-dali-cuda120>=1.43.0; extra == "nvidia-dali"
 Provides-Extra: torch-codec
-Requires-Dist: torch<=2.5.1; extra == "torch-codec"
-Requires-Dist: torchcodec>=0.3.0; extra == "torch-codec"
+Requires-Dist: torch>=2.4.1; extra == "torch-codec"
+Requires-Dist: torchcodec>=0.0.3; extra == "torch-codec"
 Provides-Extra: sklearn-datasets
 Requires-Dist: pandas>=2.2.3; extra == "sklearn-datasets"
 Requires-Dist: scikit-learn>=1.5.2; extra == "sklearn-datasets"

{sinapsis_data_readers-0.1.17 → sinapsis_data_readers-0.1.20}/src/sinapsis_data_readers.egg-info/requires.txt

@@ -35,5 +35,5 @@ sktime>=0.34.0
 soundfile>=0.12.1
 
 [torch-codec]
-torch<=2.5.1
-torchcodec>=0.3.0
+torch>=2.4.1
+torchcodec>=0.0.3