CineGuide: Conversational Movie Recommendation Assistant
Model Description
This model is a fine-tuned version of Qwen2.5-7B-Instruct for conversational movie recommendations.
Training Details
- Base Model: Qwen/Qwen2.5-7B-Instruct
- Method: LoRA (Low-Rank Adaptation) with rank-16
- Dataset: ReDial corpus (7999 training examples)
- Training Loss: 0.9140
- Perplexity: 2.49
Usage
from transformers import AutoTokenizer, AutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("./cineguide-merged")
model = AutoModelForCausalLM.from_pretrained("./cineguide-merged")
# Generate movie recommendations
prompt = "I love sci-fi movies with complex plots. Any recommendations?"
# ... [generation code]
Performance
The model shows significant improvement over the base model in:
- Providing specific movie recommendations with rationales
- Maintaining conversational context
- Understanding genre preferences
- Giving compelling explanations for recommendations
Created for CS515 Deep Learning Course Project by Serhan Yilmaz (00031275)