CoolFace
Modelpublic

pcam-interpretability/resnet50-val08933-resnet-tuned-sota

sourceHugging Faceupdated 1y agoView on Hugging Face
0likes
Model Card

resnet50

Best Validation Accuracy: 0.8933

Metadata

  • —Model Name: resnet50
  • —Optimizer: adamw
  • —Scheduler: cosine
  • —Weight Decay: 0.0005
  • —Warmup Epochs: 3
  • —Patience: 10
  • —Amp: True
  • —Seed: 42
  • —Batch Size: 32
  • —Initial Lr: 0.0005
  • —Total Epochs Ran: 23
  • —Early Stopped: True
  • —Training Time Seconds: 54549.823322057724
  • —Num Parameters: 22067009
  • —Device: NVIDIA A100-SXM4-40GB
  • —Run Id: resnet-tuned-sota

Training Configuration

  • —Epochs: 23
  • —Batch size: 32
  • —Learning rate (initial): 0.0005

Training Logs (Per Epoch)

EpochTrain LossTrain AccVal LossVal AccLR
10.32340.86240.29440.87830.000167
20.27620.88760.37290.84440.000333
30.26000.89550.42100.83330.000500
40.23250.90820.48660.81720.000500
50.21360.91690.32660.88440.000500
60.19920.92340.49220.82320.000498
70.18830.92830.36910.86510.000496
80.17940.93270.36910.87430.000492
90.17220.93540.35690.87200.000488
100.16400.93920.53300.82650.000482
110.15910.94100.31930.88540.000476
120.15430.94300.46630.85140.000469
130.14950.94440.30600.89330.000461
140.14300.94710.41980.87580.000452
150.13890.94910.34890.88850.000443
160.13470.95070.42730.87120.000432
170.13030.95210.49140.85850.000421
180.12730.95350.49110.85990.000409
190.12310.95540.42860.86580.000397
200.11900.95680.50470.85690.000384
210.11560.95790.47280.86510.000370
220.11180.95970.38970.88990.000356
230.11040.95990.57100.85130.000342