tadax / srgan Goto Github PK

These errors as follows:
File "/home/project/srgan-maset/src/train.py", line 87, in
train()
File "/home/project/srgan-maset/src/train.py", line 17, in train
model = SRGAN()
File "/home/project/srgan-maset/src/srgan.py", line 19, in init
x = pixel_shuffle_layer(x, 2, 64)
File "../utils/layer.py", line 71, in pixel_shuffle_layer
xc = tf.split(3, n_split, x)
File “ home/anaconda2/python2.7/site-packages/tensorflow/python/ops/gen_array_ops.py”, line 3426, in _split
num_split=num_split, name=name)
File "home/anaconda2/python2.7/site-pacakges/tensorflow/python/framework/op_def_library.py", line 509, in apply_op
(prefix, dtype.as_type(input_arg.type.name))
TypeError: Input 'split_dim' of 'Split' Op has type float32 that does not match except type int32.

I guess that this error is caused by using python2. Can you help me to do with this error? Thank you!
@tadax

I looked through the code and find the imported picture that are downscaled 4 times. The new picture be generated from 4 times downscaled! but i have run the whole code and i find the picture is generated from the noisy point. It is contradictive. what is my error?

It should be dlib in the Requirements instead of dilb, shouldn't it?

How to get the ImageNet? Thanks

I tried to run the train.py to train srgan. But the program terminates since there is not enough memory. My GPU is 860M with 2G memory.

How much memory exactly does the program need? Is there any way to reduce the memory needed. I tried to change the batch size but with no effect.

Thank you.

thank you for upload the code of srgan.But when I run 'python vgg19/cifar_100/preprocess.py', I got this error:
Traceback (most recent call last):
File "preprocess.py", line 50, in
data.preprocess()
File "preprocess.py", line 32, in preprocess
meta, train, test = self.load_pickle('raw')
File "preprocess.py", line 41, in load_pickle
train = pickle.load(f, encoding='latin-1')
TypeError: load() got an unexpected keyword argument 'encoding'

Can you help me?

Hi, I need to convert SMPL Meshes to SMPLX for a comparison
I am trying to setup the script in (https://github.com/vchoutas/smplx/blob/main/tools/README.md) but am facing some issues

When I run (python write_obj.py --model-folder ../models/ --motion-file ../transfer_data/support_data/github_data/amass_sample.npz --output-folder ../transfer_data/meshes/amass_sample/)

"
AttributeError: 'Struct' object has no attribute 'hands_componentsl'"

Which I know is due to some merging that needs to be done as mentioned here (https://github.com/vchoutas/smplx/blob/main/tools/README.md)

howevever those scripts generate only pkl files and and write obj script needs an npz file

how do I get the right NPZ file?

File "C:/Users/yuyan/Desktop/srgan-master/vgg19/train.py", line 71, in
train()
File "C:/Users/yuyan/Desktop/srgan-master/vgg19/train.py", line 38, in train
epoch = int(sess.run(global_step) / np.ceil(len(x_train)/batch_size)) + 1
TypeError: object of type 'NoneType' has no len()

i have trained 380 epoch with 8 gpu in three day , how much epoch should i train

Hello, I tried to modify your code to match the other things from reference paper where you deviated, but still the absurd colour artifacts visible in your results are unable to get off. Is there something else , the result of which is such colour artifacts( some pixels get absurd colour)

TypeError: integer argument expected, got float

As I still don't really understand how can i know the phi i,j as the paper described, so I searched your code and take a glimpse at your loss function & VGG definition, I don't really understand what do you mean by your phi defined in VGG build_model function, or actually say that I don't understand what does the paper mean

indicate the feature map obtained by the j-th convolution (after activation) before the i-th maxpooling layer within the VGG19 network

even I know that they want to calculate the texture content loss by feature map.

So could you plz just help me, really thanks to you any way as you have implement the paper's net

Hi Tadax, I notice that your implementation of the generator network is different from Christian's. In yours, the deconv_layer is used, while in Christian's, the convolution layer in the resnet block is used. Could you plz explain why you do in this way? Many thanks.

Orignal paper says "For each mini-batch we crop 16 random 96 × 96 HR sub images of distinct training images. Note that we can apply the generator model to images of arbitrary size as it is fully convolutional". I can see utils/augment.py but this file is no where used.

when I run "python train.py ",I got this error:

Traceback (most recent call last):
File "train.py", line 96, in
train()
File "train.py", line 17, in train
model = VGG19()
File "/home/yc/tensorflow_samples/srgan-master_tadax/vgg19/vgg19.py", line 14, in init
self.out, self.phi = self.build_model(self.x)
File "/home/yc/tensorflow_samples/srgan-master_tadax/vgg19/vgg19.py", line 23, in build_model
x = lrelu(x, self.trainable)
TypeError: lrelu() takes 1 positional argument but 2 were given

Was there anything that I have ignored? thank you