Error in server when submitting sketch.

[EmoryQ]

I have tried to set up everything per the instructions, and when I create a sketch and click "Convert Photo", I only get a spinning hourglass. The terminal says:

"FileNotFoundError: [WinError 2] The system cannot find the file specified" , but I'm not sure what file it is looking for.

This is the complete output from Powershell from the server.py:

PS C:\Users\Sequoia\smartsketch\smart-sketch\backend> python .\server.py
{'help': None, 'logging': 'info', 'log_to_stderr': None, 'log_file_prefix': None, 'log_file_max_size': 100000000, 'log_file_num_backups': 10, 'log_rotate_when': 'midnight', 'log_rotate_interval': 1, 'log_rotate_mode': 'size', 'debug': False, 'port': 80, 'address': '127.0.0.1', 'template_path': '.\\templates'}
[I 190424 11:10:56 web:2246] 200 GET / (127.0.0.1) 2.00ms
[I 190424 11:10:56 web:2246] 200 GET /static/style.css (127.0.0.1) 230.73ms
[I 190424 11:10:56 web:2246] 200 GET /static/drawingboard.min.js (127.0.0.1) 2.00ms
[I 190424 11:10:58 server:98] Recieved a file
.\img/color.png .\dataset/val_inst/greyscale.png
{(56, 79, 131): 'sea', (239, 239, 239): 'clouds', (44, 30, 22): 'dirt', (93, 110, 50): 'bush', (183, 210, 78): 'grass', (60, 59, 75): 'mountain', (152, 126, 106): 'road', (117, 158, 223): 'sky-other', (53, 38, 19): 'tree', (99, 99, 99): 'pavement', (230, 112, 182): 'flower', (193, 195, 201): 'fog', (119, 108, 45): 'hill', (191, 96, 44): 'leaves', (50, 96, 77): 'river'}
[E 190424 11:10:59 web:1788] Uncaught exception POST /upload (127.0.0.1)
    HTTPServerRequest(protocol='http', host='127.0.0.1', method='POST', uri='/upload', version='HTTP/1.1', remote_ip='127.0.0.1')
    Traceback (most recent call last):
      File "C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\site-packages\tornado\web.py", line 1697, in _execute
        result = method(*self.path_args, **self.path_kwargs)
      File ".\server.py", line 113, in post
        make_processable(greyscale_fname, output_color_file)
      File ".\server.py", line 81, in make_processable
        copy_file(ouptut_greyscale_file, ouptut_greyscale_file_labels)
      File ".\server.py", line 65, in copy_file
        subprocess.check_output(command_string.split(" "))
      File "C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\subprocess.py", line 395, in check_output
        **kwargs).stdout
      File "C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\subprocess.py", line 472, in run
        with Popen(*popenargs, **kwargs) as process:
      File "C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\subprocess.py", line 775, in __init__
        restore_signals, start_new_session)
      File "C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\subprocess.py", line 1178, in _execute_child
        startupinfo)
    FileNotFoundError: [WinError 2] The system cannot find the file specified
[E 190424 11:10:59 web:2246] 500 POST /upload (127.0.0.1) 133.84ms

[EmoryQ]

To get to the UI, after launching the server, I go to 127.0.0.1:80 in the browser.

[noyoshi]

@EmoryQ could you throw some print statements in the function copy_file so we can see what it was looking for?

[EmoryQ]

@noyoshi So I added some print statements through the code, and in the copy_file function like this:

def copy_file(old="avon.png", new="avon.png"):
    command_string = "cp " + old + " " + new
    print ("copy_file:" "cp " + old + " " + new)
    subprocess.check_output(command_string.split(" "))
    print(command_string)

The powershell output shows:

copy_file:cp dataset/val_inst/greyscale.png dataset/val_label/greyscale.png
[E 190424 18:13:49 web:1788] Uncaught exception POST /upload (127.0.0.1)

So it appears to crash the first time that the copy_file function is ran, after the command_string = "cp " + old + " " + new and before/during subprocess.check_output(command_string.split(" "))

I'm still not sure what is going on to cause this error. I'm not extremely familiar with python, and I'm not sure what the subprocess.check_output(command_string.split(" ")) command is doing.

[noyoshi]

Right here I am doing a bit of hacky stuff, I am calling the shell command cp dataset/val_inst/greyscale.png dataset/val_label/greyscale.png to move the image around to directories the model wants it in. Do you know if you have the folder dataset? And if so, dataset/val_inst and dataset/val_label? It looks like it is failing because it cannot find those folders / files.

[EmoryQ]

I do have those folders. The server does create a greyscale.png of the drawing in the val_inst folder, but it does not copy it to val_label folder.

Edit: Correction

[noyoshi]

Hmm, that is weird. I think it might be a Widows specific bug then. Later today I'll try running this on a windows machine to see if I can reproduce it. Ideally we wouldn't even be calling cp from Python - so maybe it is time that we do the file manipulation directly from Python.

[EmoryQ]

So, as a workaround: Instead of running the color_to_grey function once and then doing the copy_file, I just ran the color_to_grey function 3 times to create and put the file in the 3 locations that the copy_file function was going to send them to.

Now, however, when I click "Convert Photo" I get an "X" on the webpage where the hourglass would normally be, and I get an image of a livingroom in the /img folder (this is the same image i get when I run test.py)

[EmoryQ]

After I click "Convert to Photo", the hourglass does spin for a little bit, and the PoweShell does pause and think for a little bit. On the PowerShell, after it shows the 3 greyscale conversions it then shows:

----------------- Options ---------------
             aspect_ratio: 1.0
                batchSize: 1
      cache_filelist_read: True
     cache_filelist_write: True
          checkpoints_dir: ./checkpoints
        coco_no_portraits: False
   contain_dontcare_label: True
                crop_size: 256
                 dataroot: dataset/
             dataset_mode: coco
          display_winsize: 256
                  gpu_ids: -1
                 how_many: inf
                init_type: xavier
            init_variance: 0.02
                  isTrain: False                                [default: None]
                 label_nc: 182
       load_from_opt_file: False
                load_size: 256
         max_dataset_size: 9223372036854775807
                    model: pix2pix
                 nThreads: 0
                     name: coco_pretrained
                      nef: 16
                     netG: spade
                      ngf: 64
                  no_flip: True
              no_instance: False
                   norm_D: spectralinstance
                   norm_E: spectralinstance
                   norm_G: spectralspadesyncbatch3x3
    num_upsampling_layers: normal
                output_nc: 3
                    phase: test
          preprocess_mode: resize_and_crop
              results_dir: ./results/
           serial_batches: True
                  use_vae: False
              which_epoch: latest
                    z_dim: 256
----------------- End -------------------
dataset [CocoDataset] of size 5002 was created
Network [SPADEGenerator] was created. Total number of parameters: 97.5 million. To see the architecture, do print(network).
<torch.utils.data.dataloader.DataLoader object at 0x00000228E705C470>
torch.Size([1, 1, 256, 256])
C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\site-packages\torch\nn\modules\upsampling.py:129: UserWarning: nn.Upsample is deprecated. Use nn.functional.interpolate instead.
  warnings.warn("nn.{} is deprecated. Use nn.functional.interpolate instead.".format(self.name))
C:\Users\Sequoia\AppData\Local\Programs\Python\Python37\lib\site-packages\torch\nn\functional.py:1320: UserWarning: nn.functional.tanh is deprecated. Use torch.tanh instead.
  warnings.warn("nn.functional.tanh is deprecated. Use torch.tanh instead.")
False
C:\Users\Sequoia\smartsketch\smart-sketch\backend\img IMAGEDIR
saved image to C:\Users\Sequoia\smartsketch\smart-sketch\backend\img\8a880f3f-2b5d-49df-ae6e-306179a3196e.png
export_image_location: C:\Users\Sequoia\smartsketch\smart-sketch\backend\img\8a880f3f-2b5d-49df-ae6e-306179a3196e.png
static_image_location: C:\Users\Sequoia\smartsketch\smart-sketch\backend\img\8a880f3f-2b5d-49df-ae6e-306179a3196e.png
[I 190425 09:36:22 web:2246] 200 POST /upload (127.0.0.1) 14722.69ms
[W 190425 09:36:22 web:2246] 404 GET /C:/Users/Sequoia/smartsketch/smart-sketch/backend/img/8a880f3f-2b5d-49df-ae6e-306179a3196e.png (127.0.0.1) 0.00ms

But then I get an "X" (like missing image) on the webpage UI. I wonder what I am missing...

[EmoryQ]

It would appear that it is most likely, as you said, a windows specific bug. I'll wait for a fix, and had fun using the smartsketch.xyz when it's up and running! Thanks!!

[SeavanMao]

modify these lines and make it~

copy_file(ouptut_greyscale_file, ouptut_greyscale_file_labels)

shutil.copyfile(ouptut_greyscale_file, ouptut_greyscale_file_labels)

copy_file(ouptut_greyscale_file, ouptut_greyscale_file_img)

shutil.copyfile(ouptut_greyscale_file, ouptut_greyscale_file_img)

[parkzinc]

Thank you @SeavanMao ! im with the same problem now the server.py says :

[E 190510 08:32:05 web:1788] Uncaught exception POST /upload (127.0.0.1) HTTPServerRequest(protocol='http', host='127.0.0.1', method='POST', uri='/upload', version='HTTP/1.1', remote_ip='127.0.0.1') Traceback (most recent call last): File "C:\Users\PrkZ\AppData\Local\Programs\Python\Python36\lib\site-packages\tornado\web.py", line 1697, in _execute result = method(*self.path_args, **self.path_kwargs) File "server.py", line 115, in post make_processable(greyscale_fname, output_color_file) File "server.py", line 83, in make_processable shutil.copyfile(ouptut_greyscale_file, ouptut_greyscale_file_labels) NameError: name 'shutil' is not defined [E 190510 08:32:05 web:2246] 500 POST /upload (127.0.0.1) 155.14ms

[Dominioncher]

@SeavanMao resolve problem on windows with server error but also we need to parse path to result image. At windows we have '\\' splitter in path )

Try at this function change / splitter to \\

def parse_static_filepath(filepath): split_filepath = filepath.split('\\') while len(split_filepath) > 2: split_filepath.pop(0) return '/'.join(split_filepath)

[MyGoodSir]

@parkzinc you have to import shutil at the top of server.py after implementing all these changes it ended up working for me