#116 训练结果无法下载

结果下载中，模型的下载需要通过train_url将训练镜像的内容同步到obs中，启智平台才会提供下载。具体代码可参考示例。
从截图看日志，你的代码中是将训练镜像的内容同步到训练镜像的另一个文件夹了。可以按照以下代码进行修改：
###Copy the output model to obs###
def EnvToObs(train_dir, obs_train_url):
try:
mox.file.copy_parallel(train_dir, obs_train_url)
print("Successfully Upload {} to {}".format(train_dir,obs_train_url))
except Exception as e:
print('moxing upload {} to {} failed: '.format(train_dir,obs_train_url) + str(e))
return

###Copy the trained model data from the local running environment back to obs,
###and download it in the training task corresponding to the Qizhi platform
parser.add_argument('--train_url',
help='model folder to save/load',
default= WorkEnvironment('train') + '/model/')
train_dir = workroot + '/model'
args = parser.parse_args()
EnvToObs(train_dir, args.train_url)

结果下载没有内容，估计是没有把生成的训练结果拷贝回obs吧

建议先设置一个文件夹用来保存训练结果，训练代码那部分只能训练结果保存到前面设置的文件夹中，最后使用mox将训练结果复制到train_url这样结果查看那边就应该会有训练好的文件了。

代码示例

import moxing as mox

local_train_url = '/cache/train'

### 此处省略训练代码

mox.file.copy_parallel('local_train_url',agrs.train_url) #此处假定参数使用的是args,如果是其他的把args.替换即可

结果下载中，模型的下载需要通过train_url将训练镜像的内容同步到obs中，启智平台才会提供下载。具体代码可参考示例。
从截图看日志，你的代码中是将训练镜像的内容同步到训练镜像的另一个文件夹了。可以按照以下代码进行修改：
###Copy the output model to obs###
def EnvToObs(train_dir, obs_train_url):
try:
mox.file.copy_parallel(train_dir, obs_train_url)
print("Successfully Upload {} to {}".format(train_dir,obs_train_url))
except Exception as e:
print('moxing upload {} to {} failed: '.format(train_dir,obs_train_url) + str(e))
return

###Copy the trained model data from the local running environment back to obs,
###and download it in the training task corresponding to the Qizhi platform
parser.add_argument('--train_url',
help='model folder to save/load',
default= WorkEnvironment('train') + '/model/')
train_dir = workroot + '/model'
args = parser.parse_args()
EnvToObs(train_dir, args.train_url)

方法WorkEnvironment未定义，无法正确找到路径

Defines whether the task is a training environment or a debugging environment

def WorkEnvironment(environment):
if environment == 'train':
workroot = '/home/work/user-job-dir'
elif environment == 'debug':
workroot = '/home/work'
print('current work mode:' + environment + ', workroot:' + workroot)
return workroot

可参考示例代码https://git.openi.org.cn/OpenIOSSG/MNIST_Example/src/branch/master/train.py

修改之后依然无法下载

请检查下代码逻辑是否是将/model下的内容上传到obs，上传成功的路径如下：