suyashnpande/cinematic-music-descriptor-v2-module2

Cinematic Music Descriptor — Module 2 – Cross-Scene Context Transformer

4-layer transformer trained from scratch over sequences of Module1 scene vectors. Learns narrative arc, pacing flow, and scene-to-scene transitions.

Label Schema

Classification

• narrative_arc_position: 5 classes (Setup/Rising/Climax/Falling/Resolution)
• foreshadowing_type: 4 classes
• transition_type: 5 classes

Regression

• scene_valence_continuous: -1.0 to 1.0
• tension_level: 1–10
• arousal_level: 1–10

Training Details

• Base model: roberta-base
• Dataset: ~11,000 scenes from 60–80 movies
• Framework: PyTorch + HuggingFace Transformers
• Logging: Weights & Biases

Usage

import torch
from huggingface_hub import hf_hub_download
import config as C

# Download weights  (repo ID is built from HF_REPO_ID in config.py)
# e.g. "suyashnpande/cinematic-music-descriptor-v2-module3"
repo_id = f"{C.HF_REPO_ID}-module3"
path = hf_hub_download(repo_id=repo_id,
                       filename="module3_final.pt",
                       token=C.HF_READ_TOKEN)

Citation

If you use this model, please cite the project.