---
pid: 3194808
cwd: "/home/ubuntu/hpml_nyu"
command: "cd /home/ubuntu/hpml_nyu && source venv/bin/activate && PYTHONUNBUFFERED=1 python lab2.py --cuda --epochs 5 --task c3 2>&1"
started_at: 2026-03-05T09:38:59.425Z
running_for_seconds: 270   
---
Device: cuda
ResNet-18 params: 11,173,962

============================================================
 C3: I/O Optimization — Worker Sweep (5 epochs each)
============================================================
 Workers  AvgData(s)  AvgTrain(s)  AvgTotal(s)
       0      13.251        6.231       19.691
       4       0.625        5.361        6.403
       8       0.755        5.358        6.513
      12       0.885        5.365        6.649
      16       0.928        5.369        6.690
/home/ubuntu/.local/lib/python3.10/site-packages/torch/utils/data/dataloader.py:624: UserWarning: This DataLoader will create 20 worker processes in total. Our suggested max number of worker in current system is 16, which is smaller than what this DataLoader is going to create. Please be aware that excessive worker creation might get DataLoader running slow or even freeze, lower the worker number to avoid potential slowness/freeze if necessary.
  warnings.warn(
      20       1.051        5.362        6.821

C3.2 => Best num_workers = 4 (avg total: 6.403s, avg data: 0.625s)
Plot saved -> c3_workers.png

=> Use --num_workers 4 for subsequent experiments

---
exit_code: 0
elapsed_ms: 273576
ended_at: 2026-03-05T09:43:33.001Z
---