Tpatial-Transformer-Networks-pytorch

• Refer to daviddao/spatial-transformer-tensorflow(Tensorflow) .

• Implementation of Spatial Transformer Networks.

Statement

• Do the Experiments on the cluttered MNIST dataset of daviddao.

• The accuracy and loss records can be find in cnn.out & stn.out.

• The transform img can be find in transform_img/.

• py35_pytorch03_version contains the old version code

Environment

• python3.6

• pytorch 0.4.0

Accuracy

CNN

• Testing: epoch[195/200] loss:0.5264 acc:0.9211

• Testing: epoch[196/200] loss:0.5185 acc:0.9194

• Testing: epoch[197/200] loss:0.5160 acc:0.9158

• Testing: epoch[198/200] loss:0.5053 acc:0.9183

• Testing: epoch[199/200] loss:0.5057 acc:0.9153

STN

• Testing: epoch[195/200] loss:0.0880 acc:0.9762

• Testing: epoch[196/200] loss:0.0961 acc:0.9757

• Testing: epoch[197/200] loss:0.0893 acc:0.9742

• Testing: epoch[198/200] loss:0.1015 acc:0.9740

• Testing: epoch[199/200] loss:0.0938 acc:0.9738

Transform Image

(input|transform|input|transform)