from transformers import pipeline
import torch

# Load the model
classifier = pipeline("text-classification", 
                     model="FutureMa/game-issue-review-detection",
                     device=0 if torch.cuda.is_available() else -1)

# Define review examples
reviews = [
    "Great game ruined by the worst final boss in history. Such a slog that has to be cheesed to win.",
    "Great game, epic story, best gameplay and banger music. Overall very good jrpg games for me also i hope gallica is real"
]

# Label explanations
LABEL_MAP = {
    "LABEL_0": "Non Game Issue Review",
    "LABEL_1": "Game Issue Review"
}

# Classify and display results
print("🔍 Game Issue Review Analysis Results:\n")
print("-" * 80)
for i, review in enumerate(reviews, 1):
    pred = classifier(review)
    label_explanation = LABEL_MAP[pred[0]['label']]
    print(f"Review {i}:")
    print(f"Text: {review}")
    print(f"Classification: {label_explanation}")
    print(f"Confidence: {pred[0]['score']:.4f}")
    print("-" * 80)

Supported Languages

🌐 English

The model is particularly useful for:

Game developers monitoring player feedback
Community managers identifying trending issues
QA teams prioritizing bug fixes
Researchers analyzing game review patterns