sinapsis-data-tools 0.2.24__tar.gz → 0.2.26__tar.gz

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sinapsis-data-tools
-Version: 0.2.24
+Version: 0.2.26
 Summary: Module that contains different packages to perform data-related operations through Sinapsis templates.
 Author-email: SinapsisAI <dev@sinapsis.tech>
 Project-URL: Homepage, https://sinapsis.tech

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_analysis/src/sinapsis_data_analysis/helpers/excluded_models.py

@@ -24,3 +24,4 @@ excluded_neighbors_models = [
     "LocalOutlierFactor",
 ]
 excluded_tree_models = ["plot_tree", "export_text", "export_graphviz", "BaseDecisionTree"]
+excluded_svm_models = ["l1_min_c"]

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/ml_base_inference.py

@@ -1,4 +1,5 @@
 # -*- coding: utf-8 -*-
+import os
 from abc import abstractmethod
 from typing import Any
 
@@ -6,6 +7,7 @@ import numpy as np
 from sinapsis_core.data_containers.data_packet import DataContainer
 from sinapsis_core.template_base.base_models import TemplateAttributes
 from sinapsis_core.template_base.template import Template
+from sinapsis_core.utils.env_var_keys import SINAPSIS_CACHE_DIR
 
 
 class MLBaseInference(Template):
@@ -24,11 +26,12 @@ class MLBaseInference(Template):
         """
 
         model_path: str
+        root_dir : str = SINAPSIS_CACHE_DIR
         generic_field_key: str
 
     def __init__(self, attributes: TemplateAttributes) -> None:
         super().__init__(attributes)
-        self.model = self.load_model(self.attributes.model_path)
+
 
     def get_data(self, container: DataContainer) -> Any:
         """Get the data from the data container
@@ -111,7 +114,8 @@ class MLBaseInference(Template):
         if not self.data_is_valid(data):
             self.logger.warning("Invalid or missing data")
             return container
-
+        full_path = os.path.join(self.attributes.root_dir, self.attributes.model_path)
+        self.model = self.load_model(full_path)
         data = self.preprocess_data(data)
         predictions = self.predict(data)
 

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/ml_base_training.py

@@ -34,7 +34,7 @@ class MLBaseAttributes(TemplateAttributes):
         model_save_path (str): Path where the trained model will be saved.
     """
 
-    generic_field_key: str
+    generic_field_key: str | None = None
     root_dir : str = WORKING_DIR
     model_save_path: str
 
@@ -220,13 +220,13 @@ class MLBaseTraining(BaseDynamicWrapperTemplate):
         full_path = os.path.join(self.attributes.root_dir, self.attributes.model_save_path)
         try:
             os.makedirs(os.path.dirname(full_path), exist_ok=True)
-            self._save_model_implementation()
+            self._save_model_implementation(full_path)
             self.logger.info(f"Model saved at {self.attributes.model_save_path}")
         except (MemoryError, TypeError) as e:
             self.logger.error(f"Error saving model: {e}")
 
     @abstractmethod
-    def _save_model_implementation(self) -> None:
+    def _save_model_implementation(self, full_path: str) -> None:
         """Save the trained model using an implementation-specific method.
 
         This abstract method should be implemented by subclasses to define

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/sklearn_train.py

@@ -5,11 +5,12 @@ from sinapsis_core.template_base.base_models import UIPropertiesMetadata
 from sinapsis_core.template_base.dynamic_template import WrapperEntryConfig
 from sinapsis_core.template_base.dynamic_template_factory import make_dynamic_template
 from sinapsis_core.utils.env_var_keys import SINAPSIS_BUILD_DOCS
-from sklearn import linear_model, neighbors, neural_network, tree
+from sklearn import linear_model, neighbors, neural_network, svm, tree
 
 from sinapsis_data_analysis.helpers.excluded_models import (
     excluded_linear_models,
     excluded_neighbors_models,
+    excluded_svm_models,
     excluded_tree_models,
 )
 from sinapsis_data_analysis.helpers.tags import Tags
@@ -53,12 +54,12 @@ class SKLearnLinearModelsTrain(MLBaseTraining):
         category="SKLearn", tags=[Tags.DATA_ANALYSIS, Tags.LINEAR_REGRESSION, Tags.MODELS, Tags.SKLEARN, Tags.TRAINING]
     )
 
-    def _save_model_implementation(self) -> None:
+    def _save_model_implementation(self, full_path:str) -> None:
         """
         Implements the abstract method from the base class to
         save the model to the path specified in attributes.
         """
-        joblib.dump(self.trained_model, self.attributes.model_save_path)
+        joblib.dump(self.trained_model, full_path)
 
 
 class SKLearnNeighborsModelsTrain(SKLearnLinearModelsTrain):
@@ -177,6 +178,41 @@ class SKLearnTreeModelsTrain(SKLearnLinearModelsTrain):
     )
 
 
+class SKLearnSVMModelsTrain(SKLearnLinearModelsTrain):
+    """
+    This template dynamically wraps sklearn's svm module,
+    providing access to models like SVC, SVR, LinearSVC,
+    LinearSVR, NuSVC, NuSVR, and OneClassSVM.
+
+    Usage example:
+
+    agent:
+      name: my_test_agent
+    templates:
+    - template_name: InputTemplate
+      class_name: InputTemplate
+      attributes: {}
+    - template_name: SVCWrapper
+      class_name: SVCWrapper
+      template_input: DataLoaderTemplate
+      attributes:
+        generic_field_key: 'data_loader_key'
+        model_save_path: 'artifacts/svc_model.joblib'
+        svc_init:
+          C: 1.0
+          kernel: rbf
+          random_state: 42
+
+    """
+
+    WrapperEntry = WrapperEntryConfig(
+        wrapped_object=svm,
+        signature_from_doc_string=True,
+        exclude_module_atts=excluded_svm_models,
+        force_init_as_method=False,
+    )
+
+
 def __getattr__(name: str) -> Template:
     """
     Only create a template if it's imported, this avoids creating all the base models for all templates
@@ -190,6 +226,8 @@ def __getattr__(name: str) -> Template:
         return make_dynamic_template(name, SKLearnNNModelsTrain)
     if name in SKLearnTreeModelsTrain.WrapperEntry.module_att_names:
         return make_dynamic_template(name, SKLearnTreeModelsTrain)
+    if name in SKLearnSVMModelsTrain.WrapperEntry.module_att_names:
+        return make_dynamic_template(name, SKLearnSVMModelsTrain)
     raise AttributeError(f"template `{name}` not found in {__name__}")
 
 
@@ -198,6 +236,7 @@ __all__ = (
     + SKLearnNeighborsModelsTrain.WrapperEntry.module_att_names
     + SKLearnNNModelsTrain.WrapperEntry.module_att_names
     + SKLearnTreeModelsTrain.WrapperEntry.module_att_names
+    + SKLearnSVMModelsTrain.WrapperEntry.module_att_names
 )
 
 

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_readers/src/sinapsis_data_readers/templates/datasets_readers/sklearn_datasets.py

@@ -153,7 +153,7 @@ class SKLearnDatasets(BaseDynamicWrapperTemplate):
             y_test=pd.DataFrame(y_test),
         )
 
-        return split_data.model_dump_json(indent=2)
+        return split_data.model_dump()
 
     def execute(self, container: DataContainer) -> DataContainer:
         sklearn_dataset = self.wrapped_callable.__func__(**self.dataset_attributes.model_dump())

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_readers/src/sinapsis_data_readers/templates/datasets_readers/sktime_datasets.py

@@ -132,11 +132,11 @@ class SKTimeDatasets(BaseDynamicWrapperTemplate):
                 y_train=pd.DataFrame(y_train),
                 y_test=pd.DataFrame(y_test),
             )
-            return split_dataset.model_dump_json(indent=2)
+            return split_dataset.model_dump()
         except ValueError:
             self.logger.debug("Wrong format for split. original values")
             split_dataset = TabularDatasetSplit(x_train=pd.DataFrame(X), y_train=pd.DataFrame(y))
-            return split_dataset.model_dump_json(indent=2)
+            return split_dataset.model_dump()
 
     def create_dataset(self):
         return self.wrapped_callable.__func__(**self.dataset_attributes.model_dump())

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_readers/src/sinapsis_data_readers/templates/image_readers/base_image_folder_data_loader.py

@@ -115,7 +115,7 @@ class ImageBaseDataReader(_BaseDataReader, abc.ABC):
         Returns:
             list[str| path]: the path as string or Path object
         """
-        full_path = os.path.join(self.attributes.root_dir, self.attributes.data_dir)
+        full_path = os.path.join(self.attributes.root_dir, self.attributes.data_dir.lstrip("/"))
         data_items = [
             img_path if not return_as_str else str(img_path.resolve())
             for img_path in Path(full_path).glob(self.attributes.pattern)

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_tools.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sinapsis-data-tools
-Version: 0.2.24
+Version: 0.2.26
 Summary: Module that contains different packages to perform data-related operations through Sinapsis templates.
 Author-email: SinapsisAI <dev@sinapsis.tech>
 Project-URL: Homepage, https://sinapsis.tech

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_data_tools.egg-info/SOURCES.txt

@@ -14,7 +14,6 @@ packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/sklearn_man
 packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/sklearn_train.py
 packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/xgboost_inference.py
 packages/sinapsis_data_analysis/src/sinapsis_data_analysis/templates/xgboost_train.py
-packages/sinapsis_data_readers/src/test_gradio_client.py
 packages/sinapsis_data_readers/src/sinapsis_data_readers/__init__.py
 packages/sinapsis_data_readers/src/sinapsis_data_readers/helpers/__init__.py
 packages/sinapsis_data_readers/src/sinapsis_data_readers/helpers/coco_dataclasses.py

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_generic_data_tools/src/sinapsis_generic_data_tools/helpers/audio_encoder.py

@@ -6,7 +6,7 @@ from pydub import AudioSegment
 
 
 def audio_bytes_to_numpy(audio_bytes: bytes) -> tuple[np.ndarray, int]:
-    audio = AudioSegment.from_file(BytesIO(audio_bytes), format="mp3")
+    audio = AudioSegment.from_file(BytesIO(audio_bytes))
 
     samples = np.array(audio.get_array_of_samples())
 

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "sinapsis-data-tools"
-version = "0.2.24"
+version = "0.2.26"
 description = "Module that contains different packages to perform data-related operations through Sinapsis templates."
 authors = [{ name = "SinapsisAI", email = "dev@sinapsis.tech" }]
 dependencies = [

{sinapsis_data_tools-0.2.24 → sinapsis_data_tools-0.2.26}/packages/sinapsis_generic_data_tools/src/sinapsis_generic_data_tools/templates/mask_non_roi.py

@@ -23,8 +23,8 @@ class RegionOfInterest(BaseModel):
     roi: list[KeyPoint]
     model_config = ConfigDict(arbitrary_types_allowed=True)
 
-    @cached_property
     @computed_field
+    @cached_property
     def as_numpy_array(self) -> np.ndarray:
         """Returns the roi as a numpy array"""
         coors = [[point.x, point.y] for point in self.roi]