# -*- coding: utf-8 -*-
"""
Language Translation Dataset - Configuration File
Author: RSK World
Website: https://rskworld.in
Email: help@rskworld.in
Phone: +91 93305 39277
Copyright © 2016 RSK World. All rights reserved.

Configuration settings for the Language Translation Dataset project.
"""

from pathlib import Path

# Project Information
PROJECT_NAME = "Language Translation Dataset"
PROJECT_AUTHOR = "RSK World"
PROJECT_WEBSITE = "https://rskworld.in"
PROJECT_EMAIL = "help@rskworld.in"
PROJECT_PHONE = "+91 93305 39277"
PROJECT_YEAR = "2016"

# Directory Paths
BASE_DIR = Path(__file__).parent
DATA_DIR = BASE_DIR / 'data'
SCRIPTS_DIR = BASE_DIR / 'scripts'

# Dataset Configuration
LANGUAGES = ['english', 'spanish', 'french', 'german']
TRAIN_FILE_TSV = DATA_DIR / 'train.tsv'
TRAIN_FILE_JSON = DATA_DIR / 'train.json'
VALIDATION_FILE_TSV = DATA_DIR / 'validation.tsv'
VALIDATION_FILE_JSON = DATA_DIR / 'validation.json'
SAMPLE_FILE_JSON = DATA_DIR / 'sample_data.json'

# Model Configuration (for future use)
MODEL_NAMES = {
    'mbert': 'bert-base-multilingual-cased',
    'mt5': 'google/mt5-small',
    'mbart': 'facebook/mbart-large-50'
}

# Training Configuration (for future use)
MAX_LENGTH = 512
BATCH_SIZE = 16
LEARNING_RATE = 2e-5
NUM_EPOCHS = 3