how to get d_srgan.npz load

[wl3b10s]

when using training stage.

i got "d_srgan.npz" not found.

there is g_srgan download link in readme, but how to get d_srgan.npz??

thanks.

[vivid-chen]

you can annotation load .npz # tl.files.load_and_assign_npz(sess=sess, name=checkpoint_dir + '/d_{}.npz'.format(tl.global_flag['mode']), network=net_d)

and run main.py d_srgan.npz will be created.

[zhaoyucong]

@wl3b10s Can you train normally? I have the same problem

[vivid-chen]

at first， you should comment the code(#......) about loading the file d_srgan.npz. then you can try to run the code of training. and a d_srgan.npz will be generated by the code after the first training.

PS:sorry, my english is poor and you may understand what i mean?

------------------ 原始邮件 ------------------ 发件人: "zhaoyucong"[email protected]; 发送时间: 2019年5月14日(星期二) 下午4:23 收件人: "tensorlayer/srgan"[email protected]; 抄送: "陈智勇"[email protected];"Comment"[email protected]; 主题: Re: [tensorlayer/srgan] how to get d_srgan.npz load (#116)

@wl3b10s Can you train normally? I have the same problem

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

[zhaoyucong]

@vivid-chen I commented the code as you said, but he still can't train. Here are his closing tips tensorflow 1.4.1 tensorlayer 1.8.1

2019-05-15 16:12:17.332395: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1080 Ti, pci bus id: 0000:1a:00.0, compute capability: 6.1) [TL] [*] Load checkpoint/g_srgan.npz SUCCESS! Loading conv1_1: (3, 3, 3, 64), (64,) Loading conv1_2: (3, 3, 64, 64), (64,) Loading conv2_1: (3, 3, 64, 128), (128,) Loading conv2_2: (3, 3, 128, 128), (128,) Loading conv3_1: (3, 3, 128, 256), (256,) Loading conv3_2: (3, 3, 256, 256), (256,) Loading conv3_3: (3, 3, 256, 256), (256,) Loading conv3_4: (3, 3, 256, 256), (256,) Loading conv4_1: (3, 3, 256, 512), (512,) Loading conv4_2: (3, 3, 512, 512), (512,) Loading conv4_3: (3, 3, 512, 512), (512,) Loading conv4_4: (3, 3, 512, 512), (512,) Loading conv5_1: (3, 3, 512, 512), (512,) Loading conv5_2: (3, 3, 512, 512), (512,) Loading conv5_3: (3, 3, 512, 512), (512,) Loading conv5_4: (3, 3, 512, 512), (512,) Loading fc6: (25088, 4096), (4096,) Loading fc7: (4096, 4096), (4096,) Loading fc8: (4096, 1000), (1000,) sample HR sub-image: (16, 384, 384, 3) -1.0 1.0 sample LR sub-image: (16, 96, 96, 3) -1.0 1.0 ** fixed learning rate: 0.000100 (for init G) 2019-05-15 16:12:31.902147: E tensorflow/stream_executor/cuda/cuda_dnn.cc:378] Loaded runtime CuDNN library: 7103 (compatibility version 7100) but source was compiled with 7005 (compatibility version 7000). If using a binary install, upgrade your CuDNN library to match. If building from sources, make sure the library loaded at runtime matches a compatible version specified during compile configuration. 2019-05-15 16:12:31.903002: F tensorflow/core/kernels/conv_ops.cc:667] Check failed: stream->parent()->GetConvolveAlgorithms( conv_parameters.ShouldIncludeWinogradNonfusedAlgo<T>(), &algorithms) Aborted (core dumped)

[Qixiaojiu]

@zhaoyucong Can you train normally? I have the same problem. And I can't find Vgg19.npz.