Unet_binary segmentation

[jmstadt]

Agree, this is very helpful, got what I needed to work for Segmenting some cells. Did anyone try and deploy a serving of this? I can get it working locally on my Flask application but just including the above:

class SegLabelListCustom(SegmentationLabelList):
def open(self, fn): return open_mask(fn, div=True)

class SegItemListCustom(SegmentationItemList):
_label_cls = SegLabelListCustom

In my Flask app prior to loading the .pkl:

learn = load_learner(path, 'cell_export.pkl')

But when I try and deploy on Render (using Gunicorn), the same Flask app gives the following error:

  File "/opt/render/project/src/.venv/bin/gunicorn", line 10, in <module>
    sys.exit(run())
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/app/wsgiapp.py", line 61, in run
    WSGIApplication("%(prog)s [OPTIONS] [APP_MODULE]").run()
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/app/base.py", line 223, in run
    super(Application, self).run()
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/app/base.py", line 72, in run
    Arbiter(self).run()
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/arbiter.py", line 60, in __init__
    self.setup(app)
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/arbiter.py", line 120, in setup
    self.app.wsgi()
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/app/base.py", line 67, in wsgi
    self.callable = self.load()
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/app/wsgiapp.py", line 52, in load
    return self.load_wsgiapp()
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/app/wsgiapp.py", line 41, in load_wsgiapp
    return util.import_app(self.app_uri)
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/gunicorn/util.py", line 350, in import_app
    __import__(module)
  File "/opt/render/project/src/app.py", line 49, in <module>
    learn = load_learner(path, export_file_name)
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/fastai/basic_train.py", line 598, in load_learner
    state = torch.load(source, map_location='cpu') if defaults.device == torch.device('cpu') else torch.load(source)
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/torch/serialization.py", line 387, in load
    return _load(f, map_location, pickle_module, **pickle_load_args)
  File "/opt/render/project/src/.venv/lib/python3.7/site-packages/torch/serialization.py", line 574, in _load
    result = unpickler.load()
AttributeError: Can't get attribute 'SegItemListCustom' on <module '__main__' from '/opt/render/project/src/.venv/bin/gunicorn'>

It seems like Gunicorn is not reading the custom classes. I have been searching StackOverflow but have not found an answer.

[harikrishnanrajeev (hari rajeev)]

has anybody tried comparing U-NET with Mask-RCNN ?. In fastai , why do we use U-NET for segmentation ?.

There is a paper which talks about an ensemble of U-NET and Mask-RCNN

[digitalspecialists (RobG)]

Performance edge. Have a look at the kaggle data science bowl solutions to compare. Ensembles of unet/maskrcnn in this and other segmentation challenges (carvana, airbus, etc) haven’t been top performing. https://www.kaggle.com/c/data-science-bowl-2018/discussion

[mehrangh66 (Mehran Ghandehari)]

I am trying the following code:
class SegLabelListCustom(SegmentationLabelList):
def open(self, fn): return open_mask(fn, div=True)

class SegItemListCustom(SegmentationLabelList):
_label_cls = SegLabelListCustom

src = (SegItemListCustom.from_folder(path_img)
.split_by_idx(valid_idx)
.label_from_func(get_y_fn, classes=codes))

and get the following error:

Traceback (most recent call last):
File “C:\Users\Mehran\Anaconda2\envs\test2\lib\site-packages\IPython\core\interactiveshell.py”, line 3296, in run_code
exec(code_obj, self.user_global_ns, self.user_ns)
File “”, line 11, in
.label_from_func(get_y_fn, classes=codes))
File “C:\Users\Mehran\Anaconda2\envs\test2\lib\site-packages\fastai\data_block.py”, line 468, in _inner
self.process()
File “C:\Users\Mehran\Anaconda2\envs\test2\lib\site-packages\fastai\data_block.py”, line 522, in process
for ds,n in zip(self.lists, [‘train’,‘valid’,‘test’]): ds.process(xp, yp, name=n)
File “C:\Users\Mehran\Anaconda2\envs\test2\lib\site-packages\fastai\data_block.py”, line 699, in process
self.x.process(xp)
File “C:\Users\Mehran\Anaconda2\envs\test2\lib\site-packages\fastai\data_block.py”, line 75, in process
for p in self.processor: p.process(self)
File “C:\Users\Mehran\Anaconda2\envs\test2\lib\site-packages\fastai\vision\data.py”, line 372, in process
def process(self, ds:ItemList): ds.classes,ds.c = self.classes,len(self.classes)
TypeError: object of type ‘NoneType’ has no len()

[rors101 (rors)]

Im also running in to some headaches here,

See attached screenshot. Now im wondering if it’s better to just try and convert into 0s and 1s?

[Mghari (med Mghari)]

HI @bluesky314, I have the same issue. can you please put how you defined ‘src’ and ‘data’ variables? as Jeremy named them.

[xnet]

Does this work with dataframes? I applied all the fixes.

class SegLabelListCustom(SegmentationLabelList):
def open(self, fn): return open_mask(fn, div=True)

class SegItemListCustom(SegmentationItemList):
_label_cls = SegLabelListCustom

src = (SegItemListCustom.from_df(trnval_df, path=path, cols=‘filename_x’)
.split_none()
.label_from_df(cols=‘filename_y’))

I am getting:

~/anaconda3/envs//lib/python3.6/site-packages/fastai/data_block.py in process(self)
    529         "Process the inner datasets."
    530         xp,yp = self.get_processors()
--> 531         for ds,n in zip(self.lists, ['train','valid','test']): ds.process(xp, yp, name=n)
    532         #progress_bar clear the outputs so in some case warnings issued during processing disappear.
    533         for ds in self.lists:

~/anaconda3/envs//lib/python3.6/site-packages/fastai/data_block.py in process(self, xp, yp, name)
    694     def process(self, xp:PreProcessor=None, yp:PreProcessor=None, name:str=None):
    695         "Launch the processing on `self.x` and `self.y` with `xp` and `yp`."
--> 696         self.y.process(yp)
    697         if getattr(self.y, 'filter_missing_y', False):
    698             filt = array([o is None for o in self.y.items])

~/anaconda3/envs//lib/python3.6/site-packages/fastai/data_block.py in process(self, processor)
     81         if processor is not None: self.processor = processor
     82         self.processor = listify(self.processor)
---> 83         for p in self.processor: p.process(self)
     84         return self
     85 

~/anaconda3/envs//lib/python3.6/site-packages/fastai/vision/data.py in process(self, ds)
    370     "`PreProcessor` that stores the classes for segmentation."
    371     def __init__(self, ds:ItemList): self.classes = ds.classes
--> 372     def process(self, ds:ItemList):  ds.classes,ds.c = self.classes,len(self.classes)
    373 
    374 class SegmentationLabelList(ImageList):

TypeError: object of type 'NoneType' has no len()

If I call the code without .label_from_df(), it works, but I can’t get the labels. The cols in my df are basically the filepath to the images. My masks are 3-channel JPEG, binary mask with numbers 0 or 255.

[harshasatyavardhan (harsha)]

I am facing this error when training segmentation, don’t know how to solve it please help @ptrampert @jeremy @sgugger

[ptrampert (Patrick Trampert)]

Add a breakpoint before the return via set_trace() and look what is in input and what in targs, as it is quite clear that the sizes are not compatible, but hard to say why without more info.

[harshasatyavardhan (harsha)]

more details about my problem, how to add a breakpoint before set_trace().sorry if my questions are too basic as i am a newbie i am finding hard to rectify the problem @ptrampert

[ptrampert (Patrick Trampert)]

As you are in a segmentation setting and your images have a size of 256 x 256, you have to predict 256*256 pixel values, which is the 65536 elements of targs. Not clear, why your input only has 512.

set_trace() so to say is your breakpoint. The program will go into debug mode from this line on and you can debug. Also have a look here:

https://docs.python.org/3/library/pdb.html

[harshasatyavardhan (harsha)]

i did what you said i am getting this. i couldn’t find the problem @ptrampert @sgugger @init_27

[al24]

Quick question:

Can someone explain (or point me to where there is an explanation) on why we need to define a new class around SegmentationImageList and SegmentationItemList for a binary segmentation problem? I am not seeing what these new classes are achieving.

[NicWick (Nicholas Wickman)]

I believe it is so we can set div=True to load the masks as a float between 0-1 instead of 0-255

[ptrampert (Patrick Trampert)]

For binary if the classes are 0 and 255 it should work with div=True, yes. But if you have 0 and e.g. 2, it won’t work.

@ harshasatyavardhan Regarding the ste_trace(), you must place it at the spot where the problem is, means after the line targs = … and before the return line in the accuracy function.

[harshasatyavardhan (harsha)]

something strange thing happening when I am training my data set I can’t predict the iou and dice is not improving and acc_carvana is static .when I see the results the segmentation is not appearing @ptrampert @daveluo @digitalspecialists @sgugger @NicWick

[nvs232 (NVS Abhilash)]

Just for future reference, ImageItemList is renamed to ImageList. https://docs.fast.ai/vision.data.html#ImageList

[harikrishnanrajeev (hari rajeev)]

you might need more data. Try augmenting.

[harshasatyavardhan (harsha)]

i have totally 600 images and 600 masks isnt it enough.its a medical dataset.but why the metrics are so different .iou and dice are showing zero and acc_carvana is static @harikrishnanrajeev

[harikrishnanrajeev (hari rajeev)]

i did face similar situation before and adding more data helped. Try adding more data or do data augmentation. How many classes do you have ?.