无法在Floydhub上运行本地培训脚本

时间:2019-06-19 11:15:16

标签: python-3.x tensorflow training-data floydhub

我是不熟悉Fluydhub和深度学习的人。我想在floyd上运行python培训脚本,但是工作失败,并且在floyd控制台上出现了一些错误。

我有一个要训练的rnn模型,我想在Floydhub上训练它。我创建了一个项目,并使用本地项目目录对其进行了初始化,创建并上传了数据集并将其安装到该项目中。但是,当我尝试在gpu tensorflow 1.0 python 3.5环境中运行训练脚本(“ chatbot.py”)时,

floyd run --gpu --env tensorflow-1.0 "python chatbot.py"

在floyd控制台中出现了一些我无法理解的错误。

2019-06-19 03:46:19,264 INFO - Preparing to run TaskInstance <TaskInstance: arjunajay95/projects/chatbot/6 (id: ZBqYtaWUdzvgnENaMZ88Tk)
2019-06-19 03:46:19,292 INFO - Starting attempt 1
2019-06-19 03:46:19,305 INFO - Downloading and setting up data sources
2019-06-19 03:46:19,962 INFO - Using Docker image: floydhub/tensorflow:1.0.1-gpu-py3.7
2019-06-19 03:46:20,175 INFO - Starting container...
2019-06-19 03:47:57,131 INFO - 
################################################################################

2019-06-19 03:47:57,133 INFO - Run Output:
2019-06-19 03:47:59,138 INFO - Starting services.
2019-06-19 03:48:11,058 INFO - Error processing line 1 of /usr/local/lib/python3.5/site-packages/matplotlib-2.0.2-py3.5-nspkg.pth:
2019-06-19 03:48:11,058 INFO - 
2019-06-19 03:48:11,485 INFO - Failed to import the site module
2019-06-19 03:48:11,486 INFO - Traceback (most recent call last):
2019-06-19 03:48:11,486 INFO - File "/usr/local/lib/python3.5/site.py", line 167, in addpackage
2019-06-19 03:48:11,486 INFO - exec(line)
2019-06-19 03:48:11,486 INFO - File "<string>", line 1, in <module>
2019-06-19 03:48:11,560 INFO - File "/usr/local/lib/python3.5/types.py", line 166, in <module>
2019-06-19 03:48:11,561 INFO - import functools as _functools
2019-06-19 03:48:11,561 INFO - File "/usr/local/lib/python3.5/functools.py", line 23, in <module>
2019-06-19 03:48:11,561 INFO - from weakref import WeakKeyDictionary
2019-06-19 03:48:11,561 INFO - File "/usr/local/lib/python3.5/weakref.py", line 12, in <module>
2019-06-19 03:48:11,562 INFO - from _weakref import (
2019-06-19 03:48:11,562 INFO - ImportError: cannot import name '_remove_dead_weakref'
2019-06-19 03:48:11,563 INFO - 
2019-06-19 03:48:11,563 INFO - During handling of the above exception, another exception occurred:
2019-06-19 03:48:11,563 INFO - 
2019-06-19 03:48:11,566 INFO - Traceback (most recent call last):
2019-06-19 03:48:11,566 INFO - File "/usr/local/lib/python3.5/site.py", line 563, in <module>
2019-06-19 03:48:11,566 INFO - main()
2019-06-19 03:48:11,567 INFO - File "/usr/local/lib/python3.5/site.py", line 550, in main
2019-06-19 03:48:11,567 INFO - known_paths = addsitepackages(known_paths)
2019-06-19 03:48:11,567 INFO - File "/usr/local/lib/python3.5/site.py", line 327, in addsitepackages
2019-06-19 03:48:11,567 INFO - addsitedir(sitedir, known_paths)
2019-06-19 03:48:11,567 INFO - File "/usr/local/lib/python3.5/site.py", line 206, in addsitedir
2019-06-19 03:48:11,567 INFO - addpackage(sitedir, name, known_paths)
2019-06-19 03:48:11,567 INFO - File "/usr/local/lib/python3.5/site.py", line 177, in addpackage
2019-06-19 03:48:11,568 INFO - import traceback
2019-06-19 03:48:11,568 INFO - File "/usr/local/lib/python3.5/traceback.py", line 5, in <module>
2019-06-19 03:48:11,568 INFO - import linecache
2019-06-19 03:48:11,568 INFO - File "/usr/local/lib/python3.5/linecache.py", line 8, in <module>
2019-06-19 03:48:11,568 INFO - import functools
2019-06-19 03:48:11,568 INFO - File "/usr/local/lib/python3.5/functools.py", line 23, in <module>
2019-06-19 03:48:11,568 INFO - from weakref import WeakKeyDictionary
2019-06-19 03:48:11,569 INFO - File "/usr/local/lib/python3.5/weakref.py", line 12, in <module>
2019-06-19 03:48:11,569 INFO - from _weakref import (
2019-06-19 03:48:11,569 INFO - ImportError: cannot import name '_remove_dead_weakref'
2019-06-19 03:48:11,696 INFO - 
################################################################################

2019-06-19 03:48:11,696 INFO - Waiting for container to complete...
2019-06-19 03:48:11,974 INFO - Job exited with status code: 1
2019-06-19 03:48:11,992 INFO - Creating data module for output...
2019-06-19 03:48:12,375 INFO - Data module created for output.
2019-06-19 03:48:12,375 INFO - Persisting data in home...
2019-06-19 03:48:13,289 INFO - Home data persisted.
2019-06-19 03:48:13,289 INFO - [failed] Task execution failed

此脚本在本机上完美执行时即可开始训练。但是当在floyd中运行时,会出现上述错误。有人可以解释一下我做错了什么吗?

PS:请告诉我是否需要在此处提供任何其他信息,以帮助解决此问题

0 个答案:

没有答案