Closed yuyadanyadan closed 8 years ago
こちらのCaffeを使っています。 ただし、Ubuntuではビルドしたことがないのでエラーが出る可能性があります。 https://github.com/lltcggie/caffe
Yes, waifu2x-caffe is required lltcggie's modified version of caffe. I've provided lltcggie's caffe and waifu2x-caffe for ubuntu. Maybe it will helpful for you https://github.com/nagadomi/waifu2x-caffe/blob/ubuntu/INSTALL-linux.md
I'm grateful! Thank you so much !
在 2016-08-16 17:41:59,"nagadomi" notifications@github.com 写道:
Yes, waifu2x-caffe is required lltcggie's customized caffe. I provide lltcggie's caffe and waifu2x-caffe for ubuntu. Maybe it will helpful for you https://github.com/nagadomi/waifu2x-caffe/blob/ubuntu/INSTALL-linux.md
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.
Hi, I use your caffe to run the waifu2x net . And I get: Check failed: status == CUDNN_STATUS_SUCCESS (3 vs. 0) CUDNN_STATUS_BAD_PARAM (#241)
在 2016-08-16 16:44:58,"lltcggie" notifications@github.com 写道:
こちらのCaffeを使っています。 ただし、Ubuntuではビルドしたことがないのでエラーが出る可能性があります。 https://github.com/lltcggie/caffe
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.
This is my net! Could you help me? Thank you so much
name: "upconv_7" layer { name: "data" type: "Data" top: "input_data" include { phase: TRAIN } transform_param { scale: 0.00392156 mirror: true mean_file: "/home/yuyadan/src/sourceCode/lltcggie-caffe/data/waifu2x_trainning_cpp/waifu2x_mean.binaryproto" } data_param { source: "/home/yuyadan/src/sourceCode/lltcggie-caffe/examples/waifu2x_trainning_cpp/waifu2x_train0_lmdb" batch_size: 10 backend: LMDB } } layer { name: "data" type: "Data" top: "input_data" include { phase: TEST } transform_param { scale: 0.00392156 mirror: true mean_file: "/home/yuyadan/src/sourceCode/lltcggie-caffe/data/waifu2x_trainning_cpp/waifu2x_mean.binaryproto" } data_param { source: "/home/yuyadan/src/sourceCode/lltcggie-caffe/examples/waifu2x_trainning_cpp/waifu2x_val0_lmdb" batch_size: 10 backend: LMDB } } layer { name: "conv1_layer" type: "Convolution" bottom: "input_data" top: "conv1" convolution_param { num_output: 16 kernel_size: 3 stride: 1 bias_filler{ type: "constant" value: 0 } weight_filler { type: "gaussian" std: 0.01 } } } layer { name: "conv1_relu_layer" type: "ReLU" bottom: "conv1" top: "conv1" relu_param { negative_slope: 0.1 } } layer { name: "conv2_layer" type: "Convolution" bottom: "conv1" top: "conv2" convolution_param { num_output: 32 kernel_size: 3 stride: 1 bias_term: false weight_filler { type: "gaussian" std: 0.01 } } } layer { name: "conv2_relu_layer" type: "ReLU" bottom: "conv2" top: "conv2" relu_param { negative_slope: 0.1 } } layer { name: "conv3_layer" type: "Convolution" bottom: "conv2" top: "conv3" convolution_param { num_output: 64 kernel_size: 3 stride: 1 bias_term: false weight_filler { type: "gaussian" std: 0.01 } } } layer { name: "conv3_relu_layer" type: "ReLU" bottom: "conv3" top: "conv3" relu_param { negative_slope: 0.1 } } layer { name: "conv4_layer" type: "Convolution" bottom: "conv3" top: "conv4" convolution_param { num_output: 128 kernel_size: 3 stride: 1 bias_term: false weight_filler { type: "gaussian" std: 0.01 } } } layer { name: "conv4_relu_layer" type: "ReLU" bottom: "conv4" top: "conv4" relu_param { negative_slope: 0.1 } } layer { name: "conv5_layer" type: "Convolution" bottom: "conv4" top: "conv5" convolution_param { num_output: 128 kernel_size: 3 stride: 1 bias_term: false weight_filler { type: "gaussian" std: 0.01 } } } layer { name: "conv5_relu_layer" type: "ReLU" bottom: "conv5" top: "conv5" relu_param { negative_slope: 0.1 } } layer { name: "conv6_layer" type: "Convolution" bottom: "conv5" top: "conv6" convolution_param { num_output: 256 kernel_size: 3 stride: 1 bias_term: false weight_filler { type: "gaussian" std: 0.01 } } } layer { name: "conv6_relu_layer" type: "ReLU" bottom: "conv6" top: "conv6" relu_param { negative_slope: 0.1 } } layer { name: "conv7_layer" type: "Deconvolution" bottom: "conv6" top: "conv7" convolution_param { num_output: 3 kernel_size: 4 stride: 2 pad: 3 bias_term: false weight_filler { type: "gaussian" std: 0.01 } } param { lr_mult: 0 decay_mult: 0 } } layer { name: "target" type: "Data" top: "target" transform_param{ scale: 0.00392156 mirror: true mean_file: "/home/yuyadan/src/sourceCode/lltcggie-caffe/data/waifu2x_trainning_cpp/waifu2x_mean_out.binaryproto" } data_param { source: "/home/yuyadan/src/sourceCode/lltcggie-caffe/examples/waifu2x_trainning_cpp/waifu2x_train1_lmdb" batch_size: 10 backend: LMDB } include: { phase: TRAIN } } layer { name: "target" type: "Data" top: "target" transform_param{ scale: 0.00392156 mirror: true mean_file: "/home/yuyadan/src/sourceCode/lltcggie-caffe/data/waifu2x_trainning_cpp/waifu2x_mean_out.binaryproto" } data_param { source: "/home/yuyadan/src/sourceCode/lltcggie-caffe/examples/waifu2x_trainning_cpp/waifu2x_val1_lmdb" batch_size: 10 backend: LMDB } include: { phase: TEST } } layer { name: "loss" type: "EuclideanLoss" bottom: "conv7" bottom: "target" top: "loss"
}
在 2016-08-16 16:44:58,"lltcggie" notifications@github.com 写道:
こちらのCaffeを使っています。 ただし、Ubuntuではビルドしたことがないのでエラーが出る可能性があります。 https://github.com/lltcggie/caffe
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or mute the thread.
Hi, I build a project with your code in ubuntu14.04 with Nsight Eclipse. Error: my caffe::Caffe don't have SetGetcuDNNAlgorithmFunc and SetSetcuDNNAlgorithmFunc. Are they your own functions? Hoping for your reply! Thank you !