Trouble running cifar example

sean_pedersen · November 4, 2018, 1:45am

Hey Guys,
I am learning PyTorch atm and wanted to get my hands dirty with fastai as well.

So I figured just conda install like in the installation section described and I’d be able to run the examples but that did not work out (imports didn’t work). Anyway I ended up after some googling with the fastai git clone dev installation method. The imports of the example notebook examples/cifar.ipynb are running fine now, but I get the following error when running the 4th cell

learn = Learner(data, wrn_22(), metrics=accuracy).to_fp16()
learn.fit_one_cycle(30, 3e-3, wd=0.4, div_factor=10, pct_start=0.5)

The exception message says:

Exception in thread Thread-4:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range

and that a number of times.

I hope someone can guide me to fix this issue. Looking forward to explore the PyTorch / FastAI universe!

sgugger · November 4, 2018, 2:28pm

What’s your OS? It looks like pytorch v1 isn’t properly installed from your error.

sean_pedersen · November 4, 2018, 7:27pm

I am running Arch Linux on the machine producing the error. Any idea what I could try to fix the issue?

sgugger · November 5, 2018, 1:15am

Could you give us the result of:

python -c 'import fastai; fastai.show_install(1)'

sean_pedersen · November 5, 2018, 4:33pm

Here it is:

=== Software === 
python version  : 3.6.5
fastai version  : 1.0.20.dev0
torch version   : 1.0.0.dev20181029
nvidia driver   : 410.66
torch cuda ver  : 9.2.148
torch cuda is   : available
torch cudnn ver : 7104
torch cudnn is  : enabled

=== Hardware === 
nvidia gpus     : 1
torch available : 1
  - gpu0        : 8116MB | GeForce GTX 1080

=== Environment === 
platform        : Linux-4.18.16-arch1-1-ARCH-x86_64-with-arch
distro          : Arch Linux  
conda env       : Unknown
python          : /home/skalar/miniconda3/bin/python
sys.path        : 
/home/skalar/miniconda3/lib/python36.zip
/home/skalar/miniconda3/lib/python3.6
/home/skalar/miniconda3/lib/python3.6/lib-dynload
/home/skalar/miniconda3/lib/python3.6/site-packages
/home/skalar/projects/fastai
/home/skalar/miniconda3/lib/python3.6/site-packages/IPython/extensions

Mon Nov  5 18:32:36 2018       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 410.66       Driver Version: 410.66       CUDA Version: 10.0     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce GTX 1080    Off  | 00000000:65:00.0 Off |                  N/A |
| 25%   29C    P0    38W / 200W |     10MiB /  8116MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

I ran the command in fastai/fastai directory

sgugger · November 6, 2018, 12:31am

Not catching anything out of the ordinary. Does the lesson 1 from course v3 work or not? Also does this example run when you remove the fp16 call?

sean_pedersen · November 17, 2018, 7:36am

Removing FP16 did not help…

I tried running lesson1-pets from course3 but ran into the same error:
Executing learn.fit_one_cycle(4)
Resulted in:
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range

Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range

Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range
Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range

Exception in thread Thread-6:
Traceback (most recent call last):
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 916, in _bootstrap_inner
self.run()
File “/home/skalar/miniconda3/lib/python3.6/threading.py”, line 864, in run
self._target(*self._args, **self._kwargs)
File “/home/skalar/miniconda3/lib/python3.6/multiprocessing/resource_sharer.py”, line 139, in _serve
signal.pthread_sigmask(signal.SIG_BLOCK, range(1, signal.NSIG))
File “/home/skalar/miniconda3/lib/python3.6/signal.py”, line 60, in pthread_sigmask
sigs_set = _signal.pthread_sigmask(how, mask)
ValueError: signal number 32 out of range

sean_pedersen · November 17, 2018, 8:02am

Is there an up to date docker container with fastai+torch for GPU available? Maybe that could help…

yurket · February 20, 2019, 9:06pm

Run into the same problem on my PC with custom fast.ai installation. Resolved by updating python interpreter version from 3.6.5 to 3.6.6.
https://bugs.python.org/issue33329

MariosOreo · March 13, 2019, 1:08am

The same problem.

Resolved by updating Python3.5 to Python3.7