texture_nets icon indicating copy to clipboard operation
texture_nets copied to clipboard

Published 20 hours ago • verified publisher icon DmitryUlyanov

More parameter about the trainning model

Open ghost opened this issue 8 years ago • 3 comments

Hi, I have used the model by your download link, the result look nice. But can you provide more input how you train this model, for example the content and style weight and also is it possible to output original ratio size photo rather than complete square output? Also if you can provide more models for test would be appreciated! Anyway, awesome job on the feed forward method.

Thanks

ghost avatar Aug 19 '16 22:08 ghost

Also I tried to train my own model, the training stop at model_50000.t7...

ghost avatar Aug 20 '16 23:08 ghost

Well, it finishes...

DmitryUlyanov avatar Aug 21 '16 06:08 DmitryUlyanov

I use the model I trained but failed to execute, but I can successfully execute test with the model you provided.BTW, I don't have the model_final.t7 in the trained result folder, so I use model.t7. Here is some output, please advise. Maybe I have a dummy folder in /train/val/dummy cause the issue?

$ th test.lua -input_image ../contentPhoto/brad_pitt.jpg -model data/checkpoints/model.t7 -image_size 100

nn.Sequential { input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> output: nn.Concat { input |-> (1): nn.Sequential { | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> (12) -> output] | (1): nn.Concat { | input | |-> (1): nn.Sequential { | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> (12) -> output] | | (1): nn.Concat { | | input | | |-> (1): nn.Sequential { | | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> (12) -> output] | | | (1): nn.Concat { | | | input | | | |-> (1): nn.Sequential { | | | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> (12) -> output] | | | | (1): nn.Concat { | | | | input | | | | |-> (1): nn.Sequential { | | | | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> (12) -> output] | | | | | (1): nn.SpatialAveragePooling(32x32, 32,32) | | | | | (2): nn.Sequential { | | | | | [input -> (1) -> (2) -> output] | | | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | | | (2): cudnn.SpatialConvolution(3 -> 8, 3x3) | | | | | } | | | | | (3): nn.SpatialBatchNormalization | | | | | (4): nn.ReLU | | | | | (5): nn.Sequential { | | | | | [input -> (1) -> (2) -> output] | | | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | | | (2): cudnn.SpatialConvolution(8 -> 8, 3x3) | | | | | } | | | | | (6): nn.SpatialBatchNormalization | | | | | (7): nn.ReLU | | | | | (8): cudnn.SpatialConvolution(8 -> 8, 1x1) | | | | | (9): nn.SpatialBatchNormalization | | | | | (10): nn.ReLU | | | | | (11): nn.SpatialUpSamplingNearest | | | | | (12): nn.SpatialBatchNormalization | | | | | } | | | | |-> (2): nn.Sequential { | | | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> output] | | | | (1): nn.SpatialAveragePooling(16x16, 16,16) | | | | (2): nn.Sequential { | | | | [input -> (1) -> (2) -> output] | | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | | (2): cudnn.SpatialConvolution(3 -> 8, 3x3) | | | | } | | | | (3): nn.SpatialBatchNormalization | | | | (4): nn.ReLU | | | | (5): nn.Sequential { | | | | [input -> (1) -> (2) -> output] | | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | | (2): cudnn.SpatialConvolution(8 -> 8, 3x3) | | | | } | | | | (6): nn.SpatialBatchNormalization | | | | (7): nn.ReLU | | | | (8): cudnn.SpatialConvolution(8 -> 8, 1x1) | | | | (9): nn.SpatialBatchNormalization | | | | (10): nn.ReLU | | | | (11): nn.SpatialBatchNormalization | | | | } | | | | ... -> output | | | | } | | | | (2): nn.Sequential { | | | | [input -> (1) -> (2) -> output] | | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | | (2): cudnn.SpatialConvolution(16 -> 16, 3x3) | | | | } | | | | (3): nn.SpatialBatchNormalization | | | | (4): nn.ReLU | | | | (5): nn.Sequential { | | | | [input -> (1) -> (2) -> output] | | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | | (2): cudnn.SpatialConvolution(16 -> 16, 3x3) | | | | } | | | | (6): nn.SpatialBatchNormalization | | | | (7): nn.ReLU | | | | (8): cudnn.SpatialConvolution(16 -> 16, 1x1) | | | | (9): nn.SpatialBatchNormalization | | | | (10): nn.ReLU | | | | (11): nn.SpatialUpSamplingNearest | | | | (12): nn.SpatialBatchNormalization | | | | } | | | |-> (2): nn.Sequential { | | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> output] | | | (1): nn.SpatialAveragePooling(8x8, 8,8) | | | (2): nn.Sequential { | | | [input -> (1) -> (2) -> output] | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | (2): cudnn.SpatialConvolution(3 -> 8, 3x3) | | | } | | | (3): nn.SpatialBatchNormalization | | | (4): nn.ReLU | | | (5): nn.Sequential { | | | [input -> (1) -> (2) -> output] | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | (2): cudnn.SpatialConvolution(8 -> 8, 3x3) | | | } | | | (6): nn.SpatialBatchNormalization | | | (7): nn.ReLU | | | (8): cudnn.SpatialConvolution(8 -> 8, 1x1) | | | (9): nn.SpatialBatchNormalization | | | (10): nn.ReLU | | | (11): nn.SpatialBatchNormalization | | | } | | | ... -> output | | | } | | | (2): nn.Sequential { | | | [input -> (1) -> (2) -> output] | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | (2): cudnn.SpatialConvolution(24 -> 24, 3x3) | | | } | | | (3): nn.SpatialBatchNormalization | | | (4): nn.ReLU | | | (5): nn.Sequential { | | | [input -> (1) -> (2) -> output] | | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | | (2): cudnn.SpatialConvolution(24 -> 24, 3x3) | | | } | | | (6): nn.SpatialBatchNormalization | | | (7): nn.ReLU | | | (8): cudnn.SpatialConvolution(24 -> 24, 1x1) | | | (9): nn.SpatialBatchNormalization | | | (10): nn.ReLU | | | (11): nn.SpatialUpSamplingNearest | | | (12): nn.SpatialBatchNormalization | | | } | | |-> (2): nn.Sequential { | | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> output] | | (1): nn.SpatialAveragePooling(4x4, 4,4) | | (2): nn.Sequential { | | [input -> (1) -> (2) -> output] | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | (2): cudnn.SpatialConvolution(3 -> 8, 3x3) | | } | | (3): nn.SpatialBatchNormalization | | (4): nn.ReLU | | (5): nn.Sequential { | | [input -> (1) -> (2) -> output] | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | (2): cudnn.SpatialConvolution(8 -> 8, 3x3) | | } | | (6): nn.SpatialBatchNormalization | | (7): nn.ReLU | | (8): cudnn.SpatialConvolution(8 -> 8, 1x1) | | (9): nn.SpatialBatchNormalization | | (10): nn.ReLU | | (11): nn.SpatialBatchNormalization | | } | | ... -> output | | } | | (2): nn.Sequential { | | [input -> (1) -> (2) -> output] | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | (2): cudnn.SpatialConvolution(32 -> 32, 3x3) | | } | | (3): nn.SpatialBatchNormalization | | (4): nn.ReLU | | (5): nn.Sequential { | | [input -> (1) -> (2) -> output] | | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | | (2): cudnn.SpatialConvolution(32 -> 32, 3x3) | | } | | (6): nn.SpatialBatchNormalization | | (7): nn.ReLU | | (8): cudnn.SpatialConvolution(32 -> 32, 1x1) | | (9): nn.SpatialBatchNormalization | | (10): nn.ReLU | | (11): nn.SpatialUpSamplingNearest | | (12): nn.SpatialBatchNormalization | | } | |-> (2): nn.Sequential { | [input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> output] | (1): nn.SpatialAveragePooling(2x2, 2,2) | (2): nn.Sequential { | [input -> (1) -> (2) -> output] | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | (2): cudnn.SpatialConvolution(3 -> 8, 3x3) | } | (3): nn.SpatialBatchNormalization | (4): nn.ReLU | (5): nn.Sequential { | [input -> (1) -> (2) -> output] | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | (2): cudnn.SpatialConvolution(8 -> 8, 3x3) | } | (6): nn.SpatialBatchNormalization | (7): nn.ReLU | (8): cudnn.SpatialConvolution(8 -> 8, 1x1) | (9): nn.SpatialBatchNormalization | (10): nn.ReLU | (11): nn.SpatialBatchNormalization | } | ... -> output | } | (2): nn.Sequential { | [input -> (1) -> (2) -> output] | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | (2): cudnn.SpatialConvolution(40 -> 40, 3x3) | } | (3): nn.SpatialBatchNormalization | (4): nn.ReLU | (5): nn.Sequential { | [input -> (1) -> (2) -> output] | (1): nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) | (2): cudnn.SpatialConvolution(40 -> 40, 3x3) | } | (6): nn.SpatialBatchNormalization | (7): nn.ReLU | (8): cudnn.SpatialConvolution(40 -> 40, 1x1) | (9): nn.SpatialBatchNormalization | (10): nn.ReLU | (11): nn.SpatialUpSamplingNearest | (12): nn.SpatialBatchNormalization | } |-> (2): nn.Sequential { input -> (1) -> (2) -> (3) -> (4) -> (5) -> (6) -> (7) -> (8) -> (9) -> (10) -> (11) -> output: nn.SpatialAveragePooling(1x1, 1,1) (2): nn.Sequential { input -> (1) -> (2) -> output: nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) (2): cudnn.SpatialConvolution(3 -> 8, 3x3) } (3): nn.SpatialBatchNormalization (4): nn.ReLU (5): nn.Sequential { input -> (1) -> (2) -> output: nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) (2): cudnn.SpatialConvolution(8 -> 8, 3x3) } (6): nn.SpatialBatchNormalization (7): nn.ReLU (8): cudnn.SpatialConvolution(8 -> 8, 1x1) (9): nn.SpatialBatchNormalization (10): nn.ReLU (11): nn.SpatialBatchNormalization } ... -> output } (2): nn.Sequential { input -> (1) -> (2) -> output: nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) (2): cudnn.SpatialConvolution(48 -> 48, 3x3) } (3): nn.SpatialBatchNormalization (4): nn.ReLU (5): nn.Sequential { input -> (1) -> (2) -> output: nn.SpatialReplicationPadding(l=1, r=1, t=1, b=1) (2): cudnn.SpatialConvolution(48 -> 48, 3x3) } (6): nn.SpatialBatchNormalization (7): nn.ReLU (8): cudnn.SpatialConvolution(48 -> 48, 1x1) (9): nn.SpatialBatchNormalization (10): nn.ReLU (11): cudnn.SpatialConvolution(48 -> 3, 1x1) } /home/mark/torch/install/bin/lua: /home/mark/torch/install/share/lua/5.2/nn/Container.lua:67: In 1 module of nn.Sequential: In 1 module of nn.Concat: In 1 module of nn.Sequential: In 1 module of nn.Concat: In 1 module of nn.Sequential: /home/mark/torch/install/share/lua/5.2/nn/Concat.lua:27: bad argument #1 to 'copy' (sizes do not match at /home/mark/torch/extra/cutorch/lib/THC/THCTensorCopy.cu:31) stack traceback: [C]: in function 'copy' /home/mark/torch/install/share/lua/5.2/nn/Concat.lua:27: in function </home/mark/torch/install/share/lua/5.2/nn/Concat.lua:9> [C]: in function 'xpcall' /home/mark/torch/install/share/lua/5.2/nn/Container.lua:63: in function 'rethrowErrors' /home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:44: in function </home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:41> [C]: in function 'xpcall' /home/mark/torch/install/share/lua/5.2/nn/Container.lua:63: in function 'rethrowErrors' /home/mark/torch/install/share/lua/5.2/nn/Concat.lua:14: in function </home/mark/torch/install/share/lua/5.2/nn/Concat.lua:9> [C]: in function 'xpcall' /home/mark/torch/install/share/lua/5.2/nn/Container.lua:63: in function 'rethrowErrors' /home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:44: in function </home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:41> [C]: in function 'xpcall' /home/mark/torch/install/share/lua/5.2/nn/Container.lua:63: in function 'rethrowErrors' /home/mark/torch/install/share/lua/5.2/nn/Concat.lua:14: in function </home/mark/torch/install/share/lua/5.2/nn/Concat.lua:9> [C]: in function 'xpcall' /home/mark/torch/install/share/lua/5.2/nn/Container.lua:63: in function 'rethrowErrors' /home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:44: in function </home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:41> (...tail calls...) test.lua:33: in main chunk [C]: in function 'dofile' ...mark/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:145: in main chunk [C]: in ?

WARNING: If you see a stack trace below, it doesn't point to the place where this error occurred. Please use only the one above. stack traceback: [C]: in function 'error' /home/mark/torch/install/share/lua/5.2/nn/Container.lua:67: in function 'rethrowErrors' /home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:44: in function </home/mark/torch/install/share/lua/5.2/nn/Sequential.lua:41> (...tail calls...) test.lua:33: in main chunk [C]: in function 'dofile' ...mark/torch/install/lib/luarocks/rocks/trepl/scm-1/bin/th:145: in main chunk [C]: in ?

ghost avatar Aug 21 '16 07:08 ghost