Created by: trollhe

base nvidia-docker run PaddlePaddle / PaddleClas ,output this error.

error:

2020-09-08 09:14:59 INFO: classes_num : 7
2020-09-08 09:14:59 INFO: epochs : 2
2020-09-08 09:14:59 INFO: image_shape : [3, 224, 224]
2020-09-08 09:14:59 INFO: ls_epsilon : 0.1
2020-09-08 09:14:59 INFO: mode : train
2020-09-08 09:14:59 INFO: model_save_dir : ./output/
2020-09-08 09:14:59 INFO: pretrained_model : ./output/ResNet200_vd/best_model
2020-09-08 09:14:59 INFO: save_interval : 1
2020-09-08 09:14:59 INFO: topk : 2
2020-09-08 09:14:59 INFO: total_images : 1635
2020-09-08 09:14:59 INFO: use_mix : True
2020-09-08 09:14:59 INFO: valid_interval : 1
2020-09-08 09:14:59 INFO: validate : True
2020-09-08 09:14:59 ERROR: You are using paddlepaddle cpu version! Please try to install paddlepaddle-gpu to run model on GPU.
INFO 2020-09-08 01:15:03,836 utils.py:269] terminate all the procs
ERROR 2020-09-08 01:15:03,836 utils.py:413] ABORT!!! Out of all 1 trainers, the trainer process with rank=[0] was aborted. Please check its log.
INFO 2020-09-08 01:15:06,839 utils.py:269] terminate all the procs

package:

paddlepaddle         1.8.4       
paddlepaddle-gpu     1.8.4.post97

installed gpu cuda9.0,cudnn7.6.5,nvidia driver.please help ,thanks.