# NBGrader Configuration for TinyTorch ML Systems Course c = get_config() # Course Information c.CourseDirectory.course_id = "tinytorch-ml-systems" # Directory Structure c.CourseDirectory.source_directory = "modules/source" c.CourseDirectory.release_directory = "modules/release" c.CourseDirectory.submitted_directory = "modules/submitted" c.CourseDirectory.autograded_directory = "modules/autograded" c.CourseDirectory.feedback_directory = "modules/feedback" # Solution Removal Configuration c.ClearSolutions.code_stub = { "python": "# YOUR CODE HERE\nraise NotImplementedError()" } # Text Stub for written responses c.ClearSolutions.text_stub = "YOUR ANSWER HERE" c.ClearSolutions.begin_solution_delimeter = "### BEGIN SOLUTION" c.ClearSolutions.end_solution_delimeter = "### END SOLUTION" c.ClearSolutions.begin_text_delimeter = "### BEGIN TEXT SOLUTION" c.ClearSolutions.end_text_delimeter = "### END TEXT SOLUTION" # Enforce Metadata (require proper cell metadata for grading) c.ClearSolutions.enforce_metadata = True # Validation Configuration c.Validate.ignore_checksums = False # Logging Configuration c.NbGrader.log_level = "INFO" # Assignment and Student Configuration c.AssignApp.generate_unique_ids = True # Ensure uniqueness of cell IDs c.AssignApp.create_assignment = True # Autograder Configuration c.AutogradeApp.update = True # Update existing autograded notebooks c.AutogradeApp.create = True # Create new entries if they don't exist # Feedback Configuration c.FeedbackApp.generate_feedback = True c.FeedbackApp.output_dir = "modules/feedback" # Student IDs (optional: helps when testing locally or in custom workflows) c.Gradebook.db_url = "sqlite:///gradebook.db" # Hide hidden test cells in release notebooks c.ClearSolutions.remove_hidden_tests = True # Set default language for code cells (if mixed notebooks used) c.ClearSolutions.language = "python"