Utils Module (src.utils)

Utility functions and helpers.

Gym Warnings Suppression

Module: src.utils.suppress_gym_warnings

Suppress Gym deprecation warnings by redirecting to Gymnasium.

Functions

patch_gym() -> None

Redirect all ‘gym’ imports to ‘gymnasium’.

from src.utils.suppress_gym_warnings import patch_gym

# Call BEFORE importing qlib
patch_gym()

import qlib
# No more gym deprecation warnings!

Usage:

  • Call once at the start of your script

  • Must be called before importing qlib

  • Safe to call multiple times

  • Works with all qlib functionality

is_patched() -> bool

Check if gym has been patched.

from src.utils.suppress_gym_warnings import is_patched

if is_patched():
    print("Gym is patched - no warnings!")

get_effective_gym_version() -> Optional[str]

Get version of effective gym module.

from src.utils.suppress_gym_warnings import get_effective_gym_version

version = get_effective_gym_version()
print(f"Using gym version: {version}")

Context Manager

patch_gym_context()

Context manager version of patch_gym().

from src.utils.suppress_gym_warnings import patch_gym_context

with patch_gym_context():
    import qlib
    # Gym warnings suppressed within context
    qlib.init(...)

Complete Example

Qlib Script Without Warnings

#!/usr/bin/env python3
"""
Example Qlib script with gym warnings suppressed.
"""

# IMPORTANT: Patch gym BEFORE importing qlib
from src.utils.suppress_gym_warnings import patch_gym
patch_gym()

import qlib
from qlib.data import D
from qlib.contrib.model.gbdt import LGBModel
import pandas as pd

# Initialize Qlib
qlib.init(provider_uri='data/qlib/stocks_daily', region='us')

# Create model
model = LGBModel(
    learning_rate=0.05,
    n_estimators=100,
    num_leaves=31
)

# Train model (no gym warnings!)
X_train = pd.DataFrame(...)  # Your features
y_train = pd.Series(...)      # Your targets

model.fit(X_train, y_train)

print("Model trained successfully - no warnings!")

Why This Is Needed

Qlib uses Gym for its RL strategies, but Gym is deprecated in favor of Gymnasium. When importing qlib, you may see:

DeprecationWarning: The gym package is deprecated. Please use gymnasium instead.

The suppress_gym_warnings module redirects all gym imports to gymnasium, which is API-compatible and actively maintained.

All Example Scripts Use This

All example scripts in the QuantMini project use this pattern:

# examples/qlib_lgb_example.py
from src.utils.suppress_gym_warnings import patch_gym
patch_gym()

import qlib
# ... rest of example

This ensures clean output without deprecation warnings.

Technical Details

How It Works:

  1. Adds an import hook to sys.meta_path

  2. Intercepts import gym statements

  3. Redirects to import gymnasium as gym

  4. Gymnasium is API-compatible with Gym

Compatibility:

  • Works with all Qlib versions

  • Compatible with Gym and Gymnasium

  • No side effects on other imports

  • Thread-safe

Limitations:

  • Must be called before importing qlib

  • Only affects Python import system

  • Does not patch already-imported modules