superresolution

Upscale low-resolution images using sub-pixel convolutions (ESPCN)

Idle
Epoch 0/100Batch 0-
loss
-
MSE Loss
-----StepLoss
-

Hyperparameters

0.0010
100

About this example

Uses the ESPCN (Efficient Sub-Pixel Convolutional Network) architecture with Conv2d layers and PixelShuffle for learnable upscaling. The network learns to reconstruct high-frequency details from low-resolution input. PSNR (Peak Signal-to-Noise Ratio) measures reconstruction quality in decibels.

Logs
No output yet...