Add Data Science template and execution base

Makefile

@@ -100,13 +100,17 @@ start-autograder:
 
 
 # Sandbox Management
-sandbox-build-all: sandbox-build-python sandbox-build-java sandbox-build-node sandbox-build-cpp
+sandbox-build-all: sandbox-build-python sandbox-build-pyds sandbox-build-java sandbox-build-node sandbox-build-cpp
 	@echo "✅ All sandbox images built successfully!"
 
 sandbox-build-python:
 	@echo "Building Python sandbox image..."
 	docker build -t sandbox-py:latest -f sandbox_manager/images/Dockerfile.python sandbox_manager/images/
 
+sandbox-build-pyds:
+	@echo "Building Python Data Science sandbox image..."
+	docker build -t sandbox-pyds:latest -f sandbox_manager/images/Dockerfile.python-ds sandbox_manager/images/
+
 sandbox-build-java:
 	@echo "Building Java sandbox image..."
 	docker build -t sandbox-java:latest -f sandbox_manager/images/Dockerfile.java sandbox_manager/images/

README.md

@@ -221,7 +221,21 @@ Validates HTML, CSS, and JavaScript files.
 | `check_css_property` | Validate CSS rules | `selector`, `property`, `expected_value` |
 
 And much more! Check the [WebDev Template Documentation](docs/templates/web_dev.md) for the full list of tests.
-#### 4. Custom Templates
+
+#### 4. Data Science Template
+Validates data science assignments including ML model metrics, JSON/CSV outputs, and generated artifacts. Requires Python DS sandbox.
+
+| Test Name | Description | Key Parameters |
+|-----------|-------------|----------------|
+| `expect_stdout_value` | Extract a value from stdout and compare to an expected value | `extraction_pattern`, `expected_value`, `tolerance` |
+| `expect_metric` | Extract metric from stdout and validate vs threshold | `metric_pattern`, `condition`, `threshold` |
+| `expect_csv_output` | Validate generated CSV file contents | `artifact_path`, `expected_columns`, `expected_shape`, `expected_values` |
+| `expect_json_output` | Validate generated JSON structure and keys | `artifact_path`, `required_keys`, `expected_values` |
+| `expect_model_artifact` | Verify a model file was produced | `artifact_path`, `min_size_bytes` |
+
+Check the [Data Science Template Documentation](docs/template-library/data_science.md) for full details.
+
+#### 5. Custom Templates
 Upload your own test functions for specialized grading contexts:
 
 ```python

autograder/services/command_resolver.py

@@ -15,6 +15,7 @@ class CommandResolver:
     # Default command patterns for each language
     DEFAULT_COMMANDS = {
         Language.PYTHON: "python3 {filename}",
+        Language.PYTHON_DS: "python3 {filename}",
         Language.JAVA: "java {classname}",
         Language.NODE: "node {filename}",
         Language.CPP: "./{executable}",
@@ -105,7 +106,7 @@ def _auto_resolve_command(
         This is used when program_command is "CMD" placeholder.
         """
         command = None
-        if language == Language.PYTHON:
+        if language in (Language.PYTHON, Language.PYTHON_DS):
             command = f"python3 {fallback_filename}" if fallback_filename else "python3 main.py"
         elif language == Language.JAVA:
             # Java requires class name without .java extension

autograder/services/template_library_service.py

@@ -14,6 +14,7 @@
 from autograder.template_library.api_testing import ApiTestingTemplate
 from autograder.template_library.input_output import InputOutputTemplate
 from autograder.template_library.static_analysis import StaticAnalysisTemplate
+from autograder.template_library.data_science import DataScienceTemplate
 
 logger = logging.getLogger(__name__)
 
@@ -37,6 +38,7 @@ class TemplateLibraryService:
         "api": ApiTestingTemplate,
         "input_output": InputOutputTemplate,
         "static_analysis": StaticAnalysisTemplate,
+        "data_science": DataScienceTemplate,
     }
 
     def __new__(cls):

autograder/template_library/__init__.py

@@ -3,9 +3,12 @@
 from autograder.template_library.web_dev.template import WebDevTemplate
 from autograder.template_library.api_testing import ApiTestingTemplate
 from autograder.template_library.input_output import InputOutputTemplate
+from autograder.template_library.data_science import DataScienceTemplate
 
 __all__ = [
     "WebDevTemplate",
     "ApiTestingTemplate",
     "InputOutputTemplate",
+    "DataScienceTemplate",
 ]
+

autograder/template_library/api_testing.py

@@ -109,9 +109,9 @@ def execute(self, files, sandbox: SandboxContainer, endpoint: str = "", expected
                 data = response.json()
                 if data.get(expected_key) == expected_value:
                     score = 100
-                    report = t("api_testing.check_response_json.report.success", locale=locale, endpoint=endpoint, key=expected_key, value=expected_value)
+                    report = t("api_testing.check_response_json.report.success", locale=locale, endpoint=endpoint, field_key=expected_key, value=expected_value)
                 else:
-                    report = t("api_testing.check_response_json.report.failure", locale=locale, endpoint=endpoint, key=expected_key, expected=expected_value, actual=data.get(expected_key))
+                    report = t("api_testing.check_response_json.report.failure", locale=locale, endpoint=endpoint, field_key=expected_key, expected=expected_value, actual=data.get(expected_key))
             except json.JSONDecodeError:
                 report = t("api_testing.check_response_json.report.invalid_json", locale=locale, endpoint=endpoint)