SFGAN是 64*64 EuroSAT数据集上 27000图像 执行的效果很好 test样例80%
21 类 ucmerced scene classification 256*256图像 2100图像 test样例80%
修改的 SFGAN代码 执行结果
/home/gis/anaconda3/envs/pytguo35/bin/python /home/gis/PycharmProjects/guo/SFGAN-master/sfgan_train_evalnew.py
trainset shape: (256, 256, 3, 1680)
testset shape: (256, 256, 3, 420)
WARNING:tensorflow:From /home/gis/PycharmProjects/guo/SFGAN-master/sfgan_train_evalnew.py:231: softmax_cross_entropy_with_logits (from tensorflow.python.ops.nn_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Future major versions of TensorFlow will allow gradients to flow
into the labels input on backprop by default.
See tf.nn.softmax_cross_entropy_with_logits_v2.
2018-12-16 12:00:38.221510: I tensorflow/core/platform/cpu_feature_guard.cc:137] Your CPU supports instructions that this TensorFlow binary was not compiled to use: SSE4.1 SSE4.2 AVX AVX2 FMA
2018-12-16 12:00:38.297918: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:895] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero
2018-12-16 12:00:38.298185: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1105] Found device 0 with properties:
name: GeForce GTX 1080 Ti major: 6 minor: 1 memoryClockRate(GHz): 1.683
pciBusID: 0000:01:00.0
totalMemory: 10.92GiB freeMemory: 10.76GiB
2018-12-16 12:00:38.298196: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1195] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:01:00.0, compute capability: 6.1)
Epoch 0
Classifier train accuracy: 0.224
Classifier test accuracy 0.38095238095238093
Step time: 0.7341856956481934
Epoch time: 113.11737203598022
Epoch 1
Classifier train accuracy: 0.445
Classifier test accuracy 0.47619047619047616
Step time: 0.5562746524810791
Epoch time: 111.61542010307312
Epoch 2
Classifier train accuracy: 0.571
Classifier test accuracy 0.5
Step time: 0.57527756690979
Epoch time: 112.33925247192383
Epoch 3
Classifier train accuracy: 0.639
Classifier test accuracy 0.6190476190476191
Step time: 0.5988922119140625
Epoch time: 112.94010877609253
Epoch 4
Classifier train accuracy: 0.706
Classifier test accuracy 0.6666666666666666
Step time: 0.6145772933959961
Epoch time: 113.58087730407715
Epoch 5
Classifier train accuracy: 0.736
Classifier test accuracy 0.6190476190476191
Step time: 0.6346220970153809
Epoch time: 114.21302485466003
Epoch 6
Classifier train accuracy: 0.791
Classifier test accuracy 0.6904761904761905
Step time: 0.6531987190246582
Epoch time: 114.71357917785645
Epoch 7
Classifier train accuracy: 0.819
Classifier test accuracy 0.7380952380952381
Step time: 0.6749696731567383
Epoch time: 115.62474775314331
Epoch 8
Classifier train accuracy: 0.831
Classifier test accuracy 0.7380952380952381
Step time: 0.6988215446472168
Epoch time: 116.110271692276
Epoch 9
Classifier train accuracy: 0.851
Classifier test accuracy 0.7619047619047619
Step time: 0.7208564281463623
Epoch time: 116.75750017166138
Testing...
Classifier test accuracy 0.6640211640211641
Step time: 0.7208564281463623
Epoch time: 130.66029167175293
Process finished with exit code 0