H1K0
c56397df54
Runnable end-to-end report combining narrative, code, and inline figures: data and .chord format, transformer working principle, two-stage training curves, perplexity (3.58 -> 2.15), distribution-shift plot with a reading legend, qualitative examples, and a generation demo. Written in a first-person student voice. - CLAUDE.md: report is now a Jupyter notebook; GOST formatting dropped - requirements.txt: add nbconvert + ipykernel (optional, for the notebook) Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
24 lines
378 B
Plaintext
24 lines
378 B
Plaintext
# Python >= 3.11 required
|
|
# Tested on Python 3.12.10
|
|
|
|
# Core ML
|
|
torch==2.12.0
|
|
numpy==2.4.6
|
|
|
|
# Music processing
|
|
pretty_midi==0.2.11
|
|
mido==1.3.3
|
|
|
|
# Visualization
|
|
matplotlib==3.10.9
|
|
|
|
# Web UI (optional — only needed for app.py)
|
|
gradio==6.16.0
|
|
|
|
# Report notebook (optional — only needed to run notebooks/report.ipynb)
|
|
nbconvert==7.17.1
|
|
ipykernel==7.2.0
|
|
|
|
# Testing
|
|
pytest==9.0.3
|