coding_expert/config/config.py

"""
Configuration for the Coding Expert model
"""

# Core programming domains
CODING_DOMAINS = {
    "programming_languages": {
        "python": {
            "level": "expert",
            "focus": ["data structures", "algorithms", "web development", "machine learning"]
        },
        "javascript": {
            "level": "expert",
            "focus": ["frontend", "backend", "frameworks", "performance"]
        },
        "java": {
            "level": "expert",
            "focus": ["enterprise", "concurrency", "frameworks", "design patterns"]
        },
        "c++": {
            "level": "expert",
            "focus": ["systems", "performance", "templates", "memory management"]
        },
        "go": {
            "level": "expert",
            "focus": ["concurrency", "networking", "performance", "cloud"]
        }
    },
    "frameworks": {
        "web": {
            "django": "expert",
            "flask": "expert",
            "fastapi": "expert",
            "react": "expert",
            "vue": "expert",
            "angular": "expert"
        },
        "mobile": {
            "flutter": "expert",
            "react_native": "expert",
            "swift": "expert",
            "kotlin": "expert"
        },
        "cloud": {
            "aws": "expert",
            "gcp": "expert",
            "azure": "expert",
            "kubernetes": "expert"
        }
    },
    "tools": {
        "ci_cd": ["github_actions", "jenkins", "circleci", "gitlab_ci"],
        "version_control": ["git", "mercurial"],
        "package_management": ["pip", "npm", "maven", "gradle", "cargo"],
        "ide": ["vscode", "pycharm", "intellij", "vim", "emacs"]
    }
}

# Core coding tasks
CODING_TASKS = {
    "problem_solving": {
        "level": "expert",
        "subtasks": [
            "algorithm_design",
            "data_structure_selection",
            "complexity_analysis",
            "optimization"
        ]
    },
    "code_review": {
        "level": "expert",
        "subtasks": [
            "architecture_review",
            "security_review",
            "performance_review",
            "code_style_review"
        ]
    },
    "debugging": {
        "level": "expert",
        "subtasks": [
            "memory_leaks",
            "race_conditions",
            "performance_bottlenecks",
            "concurrency_issues"
        ]
    },
    "testing": {
        "level": "expert",
        "subtasks": [
            "unit_testing",
            "integration_testing",
            "performance_testing",
            "security_testing"
        ]
    },
    "architecture_design": {
        "level": "expert",
        "subtasks": [
            "microservices",
            "distributed_systems",
            "scalability",
            "fault_tolerance"
        ]
    }
}

# Core datasets
CODING_DATASETS = {
    "CodeSearchNet": {
        "source": "codeium/codeium",
        "split": "train",
        "fields": ["code", "docstring", "language", "function_name"],
        "description": "HuggingFace - multi-language code corpus",
        "tasks": ["code_search", "code_completion", "documentation"]
    },
    "HumanEval": {
        "source": "openai/human_eval",
        "split": "test",
        "fields": ["task_id", "prompt", "canonical_solution", "test", "entry_point"],
        "description": "OpenAI's functional code evaluation dataset",
        "tasks": ["code_generation", "function_implementation", "unit_testing"]
    },
    "MBPP": {
        "source": "mbpp/mbpp",
        "split": "train",
        "fields": ["task_id", "text", "code", "test_list", "challenge_test_list"],
        "description": "Mostly Basic Python Problems",
        "tasks": ["problem_solving", "code_generation", "unit_testing"]
    },
    "Spider": {
        "source": "yale-lily/spider",
        "split": "train",
        "fields": ["query", "question", "db_id", "sql"],
        "description": "Text-to-SQL mapping",
        "tasks": ["sql_generation", "text_to_sql", "database_queries"]
    },
    "DeepFix": {
        "source": "deepfix/deepfix",
        "split": "train",
        "fields": ["code", "fixed_code", "error_type"],
        "description": "Bug fixing dataset",
        "tasks": ["bug_fixing", "error_detection", "code_correction"]
    },
    "CodeXGLUE": {
        "source": "microsoft/CodeXGLUE",
        "split": "train",
        "fields": ["code", "docstring", "task", "language"],
        "description": "Multitask code understanding/generation benchmark",
        "tasks": ["code_translation", "code_summarization", "code_generation"]
    }
}

# Print configuration summary
def print_config_summary():
    print("\nCoding Expert Configuration Summary:")
    print(f"Number of domains: {len(CODING_DOMAINS)}")
    print(f"Number of languages: {len(CODING_DOMAINS['programming_languages'])}")
    print(f"Number of tasks: {len(CODING_TASKS)}")
    print(f"Number of datasets: {len(CODING_DATASETS)}")
    print("\nDataset Details:")
    for name, config in CODING_DATASETS.items():
        print(f"\n{name}:")
        print(f"Description: {config['description']}")
        print(f"Tasks: {', '.join(config['tasks'])}")
        print(f"Fields: {', '.join(config['fields'])}")

if __name__ == "__main__":
    print_config_summary()