Google Cloud ML使用对象检测模型返回空预测 [英] Google Cloud ML returns empty predictions with object detection model

问题描述

我是第一次将模型部署到Google Cloud ML.我已经在本地训练和测试了该模型，它仍然需要工作，但是可以.

I am deploying a model to Google Cloud ML for the first time. I have trained and tested the model locally and it still needs work but it works ok.

我已将其上传到Cloud ML，并使用我在本地测试的相同示例图像进行了测试，这些图像我知道会被检测到. (使用

I have uploaded it to Cloud ML and tested with the same example images I test locally that I know get detections. (using this tutorial)

执行此操作时，没有检测到任何东西.起初我以为我上传了错误的检查点，但是我进行了测试，并且相同的检查点可以离线处理这些图像，我不知道如何进一步调试.

When I do this, I get no detections. At first I thought I had uploaded the wrong checkpoint but I tested and the same checkpoint works with these images offline, I don't know how to debug further.

当我查看结果文件

prediction.results-00000-of-00001

prediction.results-00000-of-00001

只是空的

和文件

prediction.errors_stats-00000-of-00001

prediction.errors_stats-00000-of-00001

包含以下文本:(无法解码JSON对象"，1)

contains the following text: ('No JSON object could be decoded', 1)

这是检测程序未运行但未检测到任何信号的标志，还是运行时出现问题?

Is this a sign the detection has run and detected nothing, or is there some problem while running?

也许是我准备的图像上传错误?

Maybe the problem is I am preparing the images wrong for uploading?

日志完全没有错误

谢谢

我正在做更多的测试，并尝试使用命令"gcloud ml-engine局部预测"而不是通常的局部代码在本地运行模型.我得到与在线相同的结果，根本没有答案，但是也没有错误消息

I was doing more tests and tried to run the model locally using the command "gcloud ml-engine local predict" instead of the usual local code. I get the same result as online, no answer at all, but also no error message

我正在使用TF_Record文件，所以我不理解JSON响应.这是我的命令的副本:

EDIT 2: I am using a TF_Record file, so I don't understand the JSON response. Here is a copy of my command:

gcloud ml-engine作业提交预测$ {JOB_ID} --data- 格式= tf_record \ --input-paths = gs://MY_BUCKET/data_dir/inputs.tfr \ --output-path = gs://MY_BUCKET/data_dir/version4 \ --region us-central1 \ --model ="gcp_detector" \ --version ="Version4"

gcloud ml-engine jobs submit prediction ${JOB_ID} --data- format=tf_record \ --input-paths=gs://MY_BUCKET/data_dir/inputs.tfr \ --output-path=gs://MY_BUCKET/data_dir/version4 \ --region us-central1 \ --model="gcp_detector" \ --version="Version4"

推荐答案

使用以下命令

模型导出:

# From tensorflow/models
export PYTHONPATH=$PYTHONPATH:/home/[user]/repos/DeepLearning/tools/models/research:/home/[user]/repos/DeepLearning/tools/models/research/slim
cd /home/[user]/repos/DeepLearning/tools/models/research
python object_detection/export_inference_graph.py \
    --input_type encoded_image_string_tensor \
    --pipeline_config_path /home/[user]/[path]/ssd_mobilenet_v1_pets.config \
    --trained_checkpoint_prefix /[path_to_checkpoint]/model.ckpt-216593 \
    --output_directory /[output_path]/output_inference_graph.pb

云执行

gcloud ml-engine jobs submit prediction ${JOB_ID} --data-format=TF_RECORD \
    --input-paths=gs://my_inference/data_dir/inputs/* \
    --output-path=${YOUR_OUTPUT_DIR}  \
    --region us-central1 \
    --model="model_name" \
    --version="version_name" 

我不知道什么更改可以完全解决问题，但是有一些小的更改，例如tf_record现在是TF_RECORD.希望这可以帮助其他人.向Google支持人员寻求帮助的建议(他们提出了建议)

I don't know what change exactly fixes the issue, but there are some small changes like tf_record now being TF_RECORD. Hope this helps someone else. Props to google support for their help (they suggested the changes)

这篇关于Google Cloud ML使用对象检测模型返回空预测的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！