| |
| """ |
| Google Colab Environment Setup Script for EasyTranslate. |
| |
| This script is designed to be run as the first cell in a Colab notebook. |
| It handles: |
| 1. GPU verification and CUDA setup |
| 2. Repository cloning from GitHub |
| 3. Dependency installation |
| 4. Google Drive mounting |
| 5. Environment variable configuration |
| |
| Usage in Colab: |
| !wget -q https://raw.githubusercontent.com/your-org/UCAS-EasyTranslate/main/scripts/setup_colab.py |
| %run setup_colab.py |
| """ |
|
|
| import os |
| import subprocess |
| import sys |
| from pathlib import Path |
|
|
| REPO_URL = "https://github.com/your-org/UCAS-EasyTranslate.git" |
| REPO_DIR = Path("/content/UCAS-EasyTranslate") |
| DRIVE_MOUNT_POINT = "/content/drive" |
| DRIVE_BASE = "/content/drive/MyDrive/EasyTranslate" |
|
|
|
|
| def check_gpu(): |
| try: |
| import torch |
| if torch.cuda.is_available(): |
| print(f"[OK] GPU detected: {torch.cuda.get_device_name(0)}") |
| print(f" CUDA version: {torch.version.cuda}") |
| print(f" GPU memory: {torch.cuda.get_device_properties(0).total_memory / 1024**3:.1f} GB") |
| return True |
| else: |
| print("[WARN] No GPU detected. Training will be very slow.") |
| return False |
| except ImportError: |
| print("[WARN] PyTorch not found. Installing...") |
| subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "torch"]) |
| return check_gpu() |
|
|
|
|
| def clone_repository(): |
| if REPO_DIR.exists(): |
| print(f"[INFO] Repository exists at {REPO_DIR}, pulling latest...") |
| os.chdir(REPO_DIR) |
| subprocess.check_call(["git", "pull", "origin", "main"]) |
| else: |
| print(f"[INFO] Cloning repository from {REPO_URL}...") |
| subprocess.check_call(["git", "clone", REPO_URL, str(REPO_DIR)]) |
| os.chdir(REPO_DIR) |
| sys.path.insert(0, str(REPO_DIR / "src")) |
| print(f"[OK] Repository ready at {REPO_DIR}") |
|
|
|
|
| def install_dependencies(): |
| req_file = REPO_DIR / "requirements-colab.txt" |
| if req_file.exists(): |
| print("[INFO] Installing Colab dependencies...") |
| subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "-r", str(req_file)]) |
| else: |
| print("[INFO] Installing from requirements.txt...") |
| subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "-r", str(REPO_DIR / "requirements.txt")]) |
|
|
| print("[INFO] Installing EasyTranslate package...") |
| subprocess.check_call([sys.executable, "-m", "pip", "install", "-q", "-e", str(REPO_DIR)]) |
| print("[OK] All dependencies installed") |
|
|
|
|
| def mount_drive(): |
| try: |
| from google.colab import drive |
| drive.mount(DRIVE_MOUNT_POINT) |
| if os.path.exists(DRIVE_MOUNT_POINT): |
| os.makedirs(DRIVE_BASE, exist_ok=True) |
| print(f"[OK] Google Drive mounted at {DRIVE_MOUNT_POINT}") |
| print(f" Base path: {DRIVE_BASE}") |
| return True |
| except Exception as e: |
| print(f"[WARN] Google Drive mount failed: {e}") |
| return False |
|
|
|
|
| def setup_environment(): |
| os.environ["TOKENIZERS_PARALLELISM"] = "false" |
| os.environ["WANDB_MODE"] = os.environ.get("WANDB_MODE", "offline") |
| os.environ["PYTHONHASHSEED"] = "42" |
| print("[OK] Environment variables configured") |
|
|
|
|
| def main(): |
| print("=" * 60) |
| print(" EasyTranslate — Colab Environment Setup") |
| print("=" * 60) |
| print() |
|
|
| check_gpu() |
| print() |
|
|
| clone_repository() |
| print() |
|
|
| install_dependencies() |
| print() |
|
|
| mount_drive() |
| print() |
|
|
| setup_environment() |
| print() |
|
|
| print("=" * 60) |
| print(" Setup Complete! Ready for training.") |
| print("=" * 60) |
|
|
|
|
| if __name__ == "__main__": |
| main() |
