deriva-ml 1.14.39__tar.gz → 1.14.41__tar.gz

{deriva_ml-1.14.39 → deriva_ml-1.14.41}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deriva-ml
-Version: 1.14.39
+Version: 1.14.41
 Summary: Utilities to simplify use of Dervia and Pandas to create reproducable ML pipelines
 Author-email: ISRD <isrd-dev@isi.edu>
 Requires-Python: >=3.10

{deriva_ml-1.14.39 → deriva_ml-1.14.41}/src/deriva_ml/execution/workflow.py

@@ -121,6 +121,7 @@ class Workflow(BaseModel):
         if "DERIVA_ML_WORKFLOW_URL" in os.environ:
             self.url = os.environ["DERIVA_ML_WORKFLOW_URL"]
             self.checksum = os.environ["DERIVA_ML_WORKFLOW_CHECKSUM"]
+            self.git_root = Workflow._get_git_root(Path(os.environ["DERIVA_ML_NOTEBOOK_PATHL"]))
             self.is_notebook = True
 
         if not self.url:

{deriva_ml-1.14.39 → deriva_ml-1.14.41}/src/deriva_ml/run_notebook.py

@@ -76,7 +76,7 @@ class DerivaMLRunNotebookCLI(BaseCLI):
     def main(self):
         """Parse arguments and set up execution environment."""
         args = self.parse_cli()
-        notebook_file = args.notebook_file
+        notebook_file: Path = args.notebook_file
         parameter_file = args.file
 
         # args.parameter is now a list of [KEY, VALUE] lists
@@ -115,10 +115,11 @@ class DerivaMLRunNotebookCLI(BaseCLI):
                 print(f"  {param}:{value}")
             self.run_notebook(notebook_file.resolve(), parameters, kernel=args.kernel[0], log=args.log_output)
 
-    def run_notebook(self, notebook_file, parameters, kernel=None, log=False):
+    def run_notebook(self, notebook_file: Path, parameters, kernel=None, log=False):
         url, checksum = Workflow.get_url_and_checksum(Path(notebook_file))
         os.environ["DERIVA_ML_WORKFLOW_URL"] = url
         os.environ["DERIVA_ML_WORKFLOW_CHECKSUM"] = checksum
+        os.environ["DERIVA_ML_NOTEBOOK_PATH"] = notebook_file.as_posix()
 
         with tempfile.TemporaryDirectory() as tmpdirname:
             notebook_output = Path(tmpdirname) / Path(notebook_file).name

{deriva_ml-1.14.39 → deriva_ml-1.14.41}/src/deriva_ml.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: deriva-ml
-Version: 1.14.39
+Version: 1.14.41
 Summary: Utilities to simplify use of Dervia and Pandas to create reproducable ML pipelines
 Author-email: ISRD <isrd-dev@isi.edu>
 Requires-Python: >=3.10

{deriva_ml-1.14.39 → deriva_ml-1.14.41}/tests/dataset/test_download.py

@@ -1,7 +1,7 @@
 from pathlib import Path
 
 # Local imports
-from deriva_ml import DatasetSpec, DerivaML, MLVocab
+from deriva_ml import DatasetSpec, DerivaML, MLVocab, TableDefinition, VersionPart
 from deriva_ml.demo_catalog import DatasetDescription
 from tests.test_utils import MLDatasetCatalog
 
@@ -98,7 +98,6 @@ class TestDatasetDownload:
 
         self.compare_datasets(ml_instance, dataset_test, dataset_spec)
 
-
     def test_dataset_download_recurse(self, dataset_test, tmp_path):
         hostname = dataset_test.catalog.hostname
         catalog_id = dataset_test.catalog.catalog_id
@@ -147,3 +146,28 @@ class TestDatasetDownload:
 
         self.compare_datasets(ml_instance, dataset_test, current_spec)
         self.compare_datasets(ml_instance, dataset_test, new_spec)
+
+    def test_dataset_download_schemas(self, dataset_test, tmp_path):
+        hostname = dataset_test.catalog.hostname
+        catalog_id = dataset_test.catalog.catalog_id
+        ml_instance = DerivaML(hostname, catalog_id, working_dir=tmp_path, use_minid=False)
+        dataset_description = dataset_test.dataset_description
+
+        current_version = ml_instance.dataset_version(dataset_description.rid)
+        current_spec = DatasetSpec(rid=dataset_description.rid, version=current_version)
+        ml_instance.create_table(
+            TableDefinition(
+                name="NewTable",
+                column_defs=[],
+            )
+        )
+        new_version = ml_instance.increment_dataset_version(
+            dataset_rid=dataset_description.rid, component=VersionPart.minor
+        )
+        new_spec = DatasetSpec(rid=dataset_description.rid, version=new_version)
+
+        current_bag = ml_instance.download_dataset_bag(current_spec)
+        new_bag = ml_instance.download_dataset_bag(new_spec)
+
+        assert "NewTable" in new_bag.model.schemas[ml_instance.domain_schema].tables
+        assert "NewTable" not in current_bag.model.schemas[ml_instance.domain_schema].tables

{deriva_ml-1.14.39 → deriva_ml-1.14.41}/tests/execution/test_execution.py

@@ -49,7 +49,7 @@ class TestWorkflow:
         workflow_url = workflows[0]["URL"]
 
         workflow_rid = ml_instance.lookup_workflow(workflow_url)
-
+        print(f"Workflow url: {workflow_url}")
         assert workflow_url.endswith("workflow-test.py")
 
         # Make sure that workflow is not duplicated if created again.