This script trains a ConvMixer on CIFAR 10 dataset.
\nThis is not an attempt to reproduce the results of the paper. The paper uses image augmentations present in PyTorch Image Models (timm) for training. We haven't done this for simplicity - which causes our validation accuracy to drop.
\n": "\u3053\u306e\u30b9\u30af\u30ea\u30d7\u30c8\u306f\u3001CIFAR 10 \u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u3067 ConvMixer \u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u3057\u307e\u3059\u3002
\n\u3053\u308c\u306f\u8ad6\u6587\u306e\u7d50\u679c\u3092\u518d\u73fe\u3059\u308b\u8a66\u307f\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u3053\u306e\u8ad6\u6587\u3067\u306f\u3001PyTorch\u753b\u50cf\u30e2\u30c7\u30eb\uff08timm\uff09\u306b\u3042\u308b\u753b\u50cf\u62e1\u5f35\u3092\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u306b\u4f7f\u7528\u3057\u3066\u3044\u307e\u3059\u3002\u7c21\u7565\u5316\u306e\u305f\u3081\u306b\u3053\u308c\u3092\u884c\u3063\u305f\u308f\u3051\u3067\u306f\u3042\u308a\u307e\u305b\u3093\u3002\u305d\u306e\u305f\u3081\u3001\u691c\u8a3c\u306e\u7cbe\u5ea6\u304c\u4f4e\u4e0b\u3057\u307e\u3059
\u3002\n", "We use _^_0_^_ which defines all the dataset related configurations, optimizer, and a training loop.
\n": "\u30c7\u30fc\u30bf\u30bb\u30c3\u30c8\u306b\u95a2\u9023\u3059\u308b\u3059\u3079\u3066\u306e\u69cb\u6210\u3001\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc\u3001\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30eb\u30fc\u30d7\u3092\u5b9a\u7fa9\u3059\u308b\u3082\u306e\u3092\u4f7f\u7528\u3057\u3066\u3044\u307e\u3059_^_0_^_\u3002
\n", "\n": "\n", "
Create ConvMixer
\n": "ConvMixer \u306e\u4f5c\u6210
\n", "Create configurations
\n": "\u69cb\u6210\u306e\u4f5c\u6210
\n", "Create experiment
\n": "\u5b9f\u9a13\u3092\u4f5c\u6210
\n", "Do not augment images for validation
\n": "\u691c\u8a3c\u306e\u305f\u3081\u306b\u753b\u50cf\u3092\u88dc\u8db3\u3057\u306a\u3044\u3067\u304f\u3060\u3055\u3044
\n", "Kernel size of the depth-wise convolution, _^_0_^_
\n": "\u6df1\u3055\u65b9\u5411\u306e\u7573\u307f\u8fbc\u307f\u306e\u30ab\u30fc\u30cd\u30eb\u30b5\u30a4\u30ba\u3001_^_0_^_
\n", "Load configurations
\n": "\u69cb\u6210\u3092\u30ed\u30fc\u30c9
\n", "Number of ConvMixer layers or depth, _^_0_^_
\n": "ConvMixer \u30ec\u30a4\u30e4\u30fc\u306e\u6570\u307e\u305f\u306f\u6df1\u3055\u3001_^_0_^_
\n", "Number of channels in patch embeddings, _^_0_^_
\n": "\u30d1\u30c3\u30c1\u57cb\u3081\u8fbc\u307f\u306e\u30c1\u30e3\u30f3\u30cd\u30eb\u6570\u3001_^_0_^_
\n", "Number of classes in the task
\n": "\u30bf\u30b9\u30af\u5185\u306e\u30af\u30e9\u30b9\u6570
\n", "Optimizer
\n": "\u30aa\u30d7\u30c6\u30a3\u30de\u30a4\u30b6\u30fc
\n", "Set model for saving/loading
\n": "\u4fdd\u5b58/\u8aad\u307f\u8fbc\u307f\u7528\u306e\u30e2\u30c7\u30eb\u3092\u8a2d\u5b9a
\n", "Simple image augmentations
\n": "\u30b7\u30f3\u30d7\u30eb\u306a\u753b\u50cf\u88dc\u6b63
\n", "Size of a patch, _^_0_^_
\n": "\u30d1\u30c3\u30c1\u306e\u30b5\u30a4\u30ba\u3001_^_0_^_
\n", "Start the experiment and run the training loop
\n": "\u5b9f\u9a13\u3092\u958b\u59cb\u3057\u3001\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30eb\u30fc\u30d7\u3092\u5b9f\u884c\u3057\u307e\u3059
\n", "Training epochs and batch size
\n": "\u30c8\u30ec\u30fc\u30cb\u30f3\u30b0\u30a8\u30dd\u30c3\u30af\u3068\u30d0\u30c3\u30c1\u30b5\u30a4\u30ba
\n", "Train ConvMixer on CIFAR 10": "CIFAR 10 \u306e\u30c8\u30ec\u30a4\u30f3\u30b3\u30f3\u30d0\u30fc\u30df\u30ad\u30b5\u30fc" }