Hello,
I ran into this exception when trying to execute learning_transforms.py. This exception was actually caught but throw an other into ray module. Find here the full stack trace:
/home/luc/anaconda3/envs/pytorch/bin/python /home/luc/PycharmProjects/learning-circuits/learning_transforms/learning_transforms.py
WARNING: Not monitoring node memory since `psutil` is not installed. Install this with `pip install psutil` (or ray[debug]) to enable debugging of memory-related crashes.
/home/luc/PycharmProjects/learning-circuits/butterfly/butterfly_multiply.py:17: UserWarning: C++/CUDA extension isn't installed. Will use butterfly multiply implemented in Pytorch, which is much slower.
warnings.warn("C++/CUDA extension isn't installed. Will use butterfly multiply implemented in Pytorch, which is much slower.")
/home/luc/PycharmProjects/learning-circuits/butterfly/permutation_multiply.py:14: UserWarning: C++/CUDA extension isn't installed. Will use butterfly multiply implemented in Pytorch, which is much slower.
warnings.warn("C++/CUDA extension isn't installed. Will use butterfly multiply implemented in Pytorch, which is much slower.")
INFO - Transform_factorization - Running command 'run'
INFO - Transform_factorization - Started run with ID "1"
2019-11-13 09:39:15,160 WARNING worker.py:1406 -- WARNING: Not updating worker name since `setproctitle` is not installed. Install this with `pip install setproctitle` (or ray[debug]) to enable monitoring of worker processes.
2019-11-13 09:39:15,160 INFO node.py:423 -- Process STDOUT and STDERR is being redirected to /tmp/ray/session_2019-11-13_09-39-15_32391/logs.
2019-11-13 09:39:15,265 INFO services.py:363 -- Waiting for redis server at 127.0.0.1:35395 to respond...
ERROR - Transform_factorization - Failed after 0:00:00!
Traceback (most recent calls WITHOUT Sacred internals):
File "/home/luc/PycharmProjects/learning-circuits/learning_transforms/learning_transforms.py", line 176, in run
with open('../config/redis_address', 'r') as f:
FileNotFoundError: [Errno 2] No such file or directory: '../config/redis_address'
During handling of the above exception, another exception occurred:
Traceback (most recent calls WITHOUT Sacred internals):
File "/home/luc/PycharmProjects/learning-circuits/learning_transforms/learning_transforms.py", line 180, in run
ray.init()
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/worker.py", line 1473, in init
head=True, shutdown_at_exit=False, ray_params=ray_params)
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/node.py", line 103, in __init__
self.start_ray_processes()
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/node.py", line 427, in start_ray_processes
self.start_redis()
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/node.py", line 271, in start_redis
redis_max_memory=self._ray_params.redis_max_memory)
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/services.py", line 576, in start_redis
system_memory = ray.utils.get_system_memory()
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/utils.py", line 387, in get_system_memory
memory_in_bytes = vmstat("total memory") * bytes_in_kilobyte
File "/home/luc/anaconda3/envs/pytorch/lib/python3.6/site-packages/ray/utils.py", line 336, in vmstat
raise ValueError("Can't find {} in 'vmstat' output.".format(stat))
ValueError: Can't find b'total memory' in 'vmstat' output.
Process finished with exit code 1
Have you got any idea on how to solve this issue?
Thank you
Luc
It might be an error on Ray's side. Can you check that you can run Ray's quick start example (https://github.com/ray-project/ray)?
Which OS are you using? We've only tested on Linux (Ubuntu and Debian).
Hello, I ran into this exception when trying to execute
learning_transforms.py
. This exception was actually caught but throw an other intoray
module. Find here the full stack trace:Have you got any idea on how to solve this issue? Thank you Luc