An error occured on Step 4 Execute script

[w617156977]

when I execute ./pipeline,an error occurs like this:


thanks for help!

[snapfinger]

Hi, it looks like two separate errors, and the error in your second snapshot is likely to be caused by the first one.
Your second error indicates you are trying to test a model which has not yet been produced. Have you checked you have run through training procedure completely? If so, what are the files in your /data/models directory?
Also, is your machine equipped with Nvidia GPU and installed with package tensorflow-gpu (1.3.0) as stated in readme? You need these to run the code.

[w617156977]

Do I have to use all the samples for training?I didn't download all the samples because the download speed was too slow.I only used the first 16 of 82 samples.

[w617156977]

The errors occurred before the first epoch of training was completed, so no files were generated in /data/models.

[snapfinger]

Data size should not matter b/c batch size is 1. Are you using same versions of dependencies specified in readme? If everything is the same the first error should not occur.

[w617156977]

All Main Dependencies are same as that in readme ，It still doesn't work.

[w617156977]

Error occurs in Fitting model

[snapfinger]

From search results on the internet: keras-team/keras#3657, MattVitelli/GRUV#18 (comment), you may need to check if you constructed the input training numpy array correctly (i.e. not empty). Also, please paste full log output in text (not snapshot) so I can better aid you.

[w617156977]

supreme@Supreme:~/Documents/repo$ ./pipeline
data_path: /home/supreme/Documents/data
Initialization starts.
Writing training image list.
Writing testing image list.
Initialization is done.
data_path: /home/supreme/Documents/data

Creating training data...

Training data created for fold 0, plane Z
/home/supreme/anaconda2/lib/python2.7/site-packages/h5py/init.py:36: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type.
from ._conv import register_converters as _register_converters
Using TensorFlow backend.
data_path: /home/supreme/Documents/data
number of epoch: 10
learning rate: 1e-05

     Loading and preprocessing train data...

---

---

	Creating and compiling model...

---

start building NN

---

Layer (type) Output Shape Param # Connected to

input_1 (InputLayer) (None, 192, 256, 1) 0

---

conv2d_1 (Conv2D) (None, 192, 256, 64) 640 input_1[0][0]

---

conv2d_2 (Conv2D) (None, 192, 256, 64) 36928 conv2d_1[0][0]

---

max_pooling2d_1 (MaxPooling2D) (None, 96, 128, 64) 0 conv2d_2[0][0]

---

conv2d_3 (Conv2D) (None, 96, 128, 128) 73856 max_pooling2d_1[0][0]

---

conv2d_4 (Conv2D) (None, 96, 128, 128) 147584 conv2d_3[0][0]

---

max_pooling2d_2 (MaxPooling2D) (None, 48, 64, 128) 0 conv2d_4[0][0]

---

conv2d_5 (Conv2D) (None, 48, 64, 256) 295168 max_pooling2d_2[0][0]

---

conv2d_6 (Conv2D) (None, 48, 64, 256) 590080 conv2d_5[0][0]

---

max_pooling2d_3 (MaxPooling2D) (None, 24, 32, 256) 0 conv2d_6[0][0]

---

conv2d_7 (Conv2D) (None, 24, 32, 512) 1180160 max_pooling2d_3[0][0]

---

conv2d_8 (Conv2D) (None, 24, 32, 512) 2359808 conv2d_7[0][0]

---

max_pooling2d_4 (MaxPooling2D) (None, 12, 16, 512) 0 conv2d_8[0][0]

---

conv2d_9 (Conv2D) (None, 12, 16, 1024) 4719616 max_pooling2d_4[0][0]

---

conv2d_10 (Conv2D) (None, 12, 16, 512) 4719104 conv2d_9[0][0]

---

conv2d_transpose_1 (Conv2DTrans (None, 24, 32, 512) 1049088 conv2d_10[0][0]

---

concatenate_1 (Concatenate) (None, 24, 32, 1024) 0 conv2d_transpose_1[0][0]
conv2d_8[0][0]

---

conv2d_11 (Conv2D) (None, 24, 32, 512) 4719104 concatenate_1[0][0]

---

conv2d_12 (Conv2D) (None, 24, 32, 256) 1179904 conv2d_11[0][0]

---

conv2d_transpose_2 (Conv2DTrans (None, 48, 64, 256) 262400 conv2d_12[0][0]

---

concatenate_2 (Concatenate) (None, 48, 64, 512) 0 conv2d_transpose_2[0][0]
conv2d_6[0][0]

---

conv2d_13 (Conv2D) (None, 48, 64, 256) 1179904 concatenate_2[0][0]

---

conv2d_14 (Conv2D) (None, 48, 64, 128) 295040 conv2d_13[0][0]

---

conv2d_transpose_3 (Conv2DTrans (None, 96, 128, 128) 65664 conv2d_14[0][0]

---

concatenate_3 (Concatenate) (None, 96, 128, 256) 0 conv2d_transpose_3[0][0]
conv2d_4[0][0]

---

conv2d_15 (Conv2D) (None, 96, 128, 128) 295040 concatenate_3[0][0]

---

conv2d_16 (Conv2D) (None, 96, 128, 64) 73792 conv2d_15[0][0]

---

conv2d_transpose_4 (Conv2DTrans (None, 192, 256, 64) 16448 conv2d_16[0][0]

---

concatenate_4 (Concatenate) (None, 192, 256, 128 0 conv2d_transpose_4[0][0]
conv2d_2[0][0]

---

conv2d_17 (Conv2D) (None, 192, 256, 64) 73792 concatenate_4[0][0]

---

conv2d_18 (Conv2D) (None, 192, 256, 64) 36928 conv2d_17[0][0]

---

conv2d_19 (Conv2D) (None, 192, 256, 1) 65 conv2d_18[0][0]

Total params: 23,370,113
Trainable params: 23,370,113
Non-trainable params: 0

---

None

	Fitting model...

---

Epoch 1/10
Traceback (most recent call last):
File "unet.py", line 171, in
train(cur_fold, plane, batch_size, epoch, init_lr)
File "unet.py", line 166, in train
callbacks=[model_checkpoint, csv_logger])
File "/home/supreme/anaconda2/lib/python2.7/site-packages/keras/engine/training.py", line 1037, in fit
validation_steps=validation_steps)
File "/home/supreme/anaconda2/lib/python2.7/site-packages/keras/engine/training_arrays.py", line 217, in fit_loop
callbacks.on_epoch_end(epoch, epoch_logs)
File "/home/supreme/anaconda2/lib/python2.7/site-packages/keras/callbacks.py", line 77, in on_epoch_end
callback.on_epoch_end(epoch, logs)
File "/home/supreme/anaconda2/lib/python2.7/site-packages/keras/callbacks.py", line 336, in on_epoch_end
self.progbar.update(self.seen, self.log_values)
AttributeError: 'ProgbarLogger' object has no attribute 'log_values'
/home/supreme/anaconda2/lib/python2.7/site-packages/h5py/init.py:36: FutureWarning: Conversion of the second argument of issubdtype from float to np.floating is deprecated. In future, it will be treated as np.float64 == np.dtype(float).type.
from ._conv import register_converters as _register_converters
Using TensorFlow backend.
data_path: /home/supreme/Documents/data

loading model unet_fd0_Z_ep10_lr1e-5

Traceback (most recent call last):
File "testvis.py", line 181, in
test(model_to_test, cur_fold, plane, rst_path, vis)
File "testvis.py", line 65, in test
model = load_model(model_path + model_to_test + '.h5', custom_objects={'dice_coef_loss': dice_coef_loss, 'dice_coef':dice_coef})
File "/home/supreme/anaconda2/lib/python2.7/site-packages/keras/engine/saving.py", line 249, in load_model
f = h5py.File(filepath, mode='r')
File "/home/supreme/anaconda2/lib/python2.7/site-packages/h5py/_hl/files.py", line 269, in init
fid = make_fid(name, mode, userblock_size, fapl, swmr=swmr)
File "/home/supreme/anaconda2/lib/python2.7/site-packages/h5py/_hl/files.py", line 99, in make_fid
fid = h5f.open(name, flags, fapl=fapl)
File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
File "h5py/h5f.pyx", line 78, in h5py.h5f.open
IOError: Unable to open file (unable to open file: name = '/home/supreme/Documents/data/models/unet_fd0_Z_ep10_lr1e-5.h5', errno = 2, error message = 'No such file or directory', flags = 0, o_flags = 0)

[snapfinger]

Seems you didn't even sliced the data successfully (the first step in script). I just tried with two cases without problem, so data number does not matter. Check whether you put your file in correct places.

[w617156977]

Could you show me where your files are ，please?