2017-12-18 369 views
0

系統信息谷歌雲平臺培訓問題

  1. OS平臺及分銷:WINDOWS 10 64位
  2. 從(來源或二進制)TensorFlow安裝:二進制
  3. TensorFlow版本:1.4.0
  4. Python版本3.5.2(V3.5.2:4def2a2901a5,2016年6月25日,22時18分55秒)
  5. GPU:NVIDIA的GeForce 755M 2GB CPU:英特爾x64-64的Intel Core i5-4200M @ CPU 2.50Ghz,8GB內存

描述問題 我可以在本地訓練模型。準確性很好,但我想嘗試使用GCP獲得類似的結果。我跟着谷歌文檔,這裏提到:

https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/running_on_cloud.md

https://medium.com/google-cloud/object-detection-tensorflow-and-google-cloud-platform-72e0a3f3bdd6

當我嘗試使用下面的命令來運行在雲上的訓練:

gcloud ml-engine jobs submit training object_detection_188001 \ --job-dir=gs://weeddetection/train \ --packages dist/object_detection-0.1.tar.gz,slim/dist/slim-0.1.tar.gz \ --module-name object_detection.train \ --region us-central1 \ --config object_detection/samples/cloud/cloud.yml \ --train_dir=gs://weeddetection/train \ --pipeline_config_path=gs://weeddetection/data/ssd_mobilenet_v1_pets.config 

我得到以下錯誤:

ERROR: (gcloud.ml-engine.jobs.submit.training) unrecognized arguments: 
    \ 
    \ 
    \ 
    \ 
    \ 
    \ 
    --train_dir=gs://weeddetection/train 
    \ 
    --pipeline_config_path=gs://weeddetection/data/ssd_mobilenet_v1_pets.config 
weeddetection is my GC bucket name and 188001 is my shell ID. 

回答

0

如果包含自定義選項(如--train-dir--pipeline_config_path),則需要將它們放在命令末尾。您還需要使用--來分隔命令的選項和選項。試試這個:

gcloud ml-engine jobs submit training object_detection_188001 --job-dir=gs://weeddetection/train --packages dist/object_detection-0.1.tar.gz,slim/dist/slim-0.1.tar.gz --module-name object_detection.train --region us-central1 --config object_detection/samples/cloud/cloud.yml -- --train_dir=gs://weeddetection/train --pipeline_config_path=gs://weeddetection/data/ssd_mobilenet_v1_pets.config 
+0

嘿!謝謝!這次我能夠運行它。但是,我無法在張力板上監控它。 https://drive.google.com/open?id=1nHJtl5VrUEgBGkb40DJ_FFQYxC7-tbYX和https://drive.google.com/open?id=193pH6Nr9IRjhMBiI72zlc5565oBWjXUn –