google-cloud-ml-engine

Memory Leak in TensorFlow Google Cloud ML Training

由 ≡放荡痞女 提交于 2019-12-12 04:35:45
问题 I've been trying the TensorFlow tutorial scripts on Google Cloud ML. In particular I've used the cifar10 CNN tutorial scripts at https://github.com/tensorflow/models/tree/master/tutorials/image/cifar10. When I run this training script in Google Cloud ML, there is a memory leak of around 0.5% per hour. I have not made any changes to the scripts other than packaging them into the required GCP format (as described in https://cloud.google.com/ml-engine/docs/how-tos/packaging-trainer) and setting

Prediction failed: unknown error

由 对着背影说爱祢 提交于 2019-12-11 17:23:36
问题 I'm using Google Cloud Machine Learning to predict images with labels. I've trained my model, named flower and I see the API end point at Google API Exporer but, when I call the API at API Explorer, I get the following error: Image Error I can't understanding why. Thanks Ibere 回答1: I guess you followed the tutorial from https://github.com/GoogleCloudPlatform/cloudml-samples/tree/master/flowers? I had the exact same problem but with some trial and errors I succeeded with the payload: {

Error when following google object detection API tutorial

由 馋奶兔 提交于 2019-12-11 07:06:45
问题 While following the tutorial in here, I've reached the stage of training the model on the cloud. Unfortunately, the command gcloud ml-engine jobs submit training `whoami`_object_detection_`date +%s` \ --job-dir=gs://${YOUR_GCS_BUCKET}/train \ --packages dist/object_detection-0.1.tar.gz,slim/dist/slim-0.1.tar.gz \ --module-name object_detection.train \ --region us-central1 \ --config object_detection/samples/cloud/cloud.yml \ -- \ --train_dir=gs://${YOUR_GCS_BUCKET}/train \ --pipeline_config

gcloud jobs submit prediction 'can't decode json' with --data-format=TF_RECORD

由 旧街凉风 提交于 2019-12-10 21:49:45
问题 I pushed up some test data to gcloud for prediction as a binary tfrecord-file. Running my script I got the error ('No JSON object could be decoded', 162) . What do you think I am doing wrong? To push a prediction job to gcloud, i use this script: REGION=us-east1 MODEL_NAME=mymodel VERSION=v_hopt_22 INPUT_PATH=gs://mydb/test-data.tfr OUTPUT_PATH=gs://mydb/prediction.tfr JOB_NAME=pred_${MODEL_NAME}_${VERSION}_b args=" --model "$MODEL_NAME args+=" --version "$VERSION args+=" --data-format=TF

Cloud ML Engine distributed training default type for custom tf.estimator

由 守給你的承諾、 提交于 2019-12-08 11:40:27
问题 This article suggests there are three options for distributed training Data-parallel training with synchronous updates. Data-parallel training with asynchronous updates. Model-parallel training. The tutorial then goes on to suggest that the code that follows performs data-parallel training with asynchronous updates on Cloud ML Engine which behaves as "If you distribute 10,000 batches among 10 worker nodes, each node works on roughly 1,000 batches." However, it's not clear what portion of the

Google Cloud ML Engine + Tensorflow perform preprocessing/tokenization in input_fn()

由 让人想犯罪 __ 提交于 2019-12-08 08:17:30
问题 I want to perform basic preprocessing and tokenization within my input function. My data is contained in csv's in a google cloud storage bucket location (gs://) that I cannot modify. Further, I to perform any modifications on input text within my ml-engine package so that the behavior can be replicated at serving time. my input function follows the basic structure below: filename_queue = tf.train.string_input_producer(filenames) reader = tf.TextLineReader() _, rows = reader.read_up_to

How do I get online predictions in javascript for my model on Cloud Machine Learning Engine?

由 你离开我真会死。 提交于 2019-12-08 04:18:42
问题 I have successfully deployed on model on Cloud ML Engine and verified it is working with gcloud ml-engine models predict by following the instructions, now I want to send predictions to it from my web app / javascript code. How do I do that? 回答1: The online prediction API is a REST API, so you can use any library for sending HTTPS requests, although you will need to use Google's OAuth library to get your credentials. We will use the googleapis library for simplicity. The format of the

How do I get online predictions in javascript for my model on Cloud Machine Learning Engine?

由 眉间皱痕 提交于 2019-12-08 03:30:26
I have successfully deployed on model on Cloud ML Engine and verified it is working with gcloud ml-engine models predict by following the instructions , now I want to send predictions to it from my web app / javascript code. How do I do that? The online prediction API is a REST API, so you can use any library for sending HTTPS requests, although you will need to use Google's OAuth library to get your credentials. We will use the googleapis library for simplicity. The format of the prediction request is JSON, as described in the docs . To exemplify, consider the Census example. A client for

Locally load saved tensorflow model .pb from google cloud machine learning engine

由 拟墨画扇 提交于 2019-12-06 13:57:04
问题 I'd like to take the tensorflow model i've trained online and run it locally with a python program I distribute. After training, I get a directory /model with two files /saved_model.pb and a folder /variables. What is the simplest way to deploy this locally? I was following here for deploying frozen models, but I can't quite read in the .pb. I downloaded saved_model.pb to my working directly and tried with tf.gfile.GFile("saved_model.pb", "rb") as f: graph_def = tf.GraphDef() graph_def

Locally load saved tensorflow model .pb from google cloud machine learning engine

由 岁酱吖の 提交于 2019-12-04 20:08:59
I'd like to take the tensorflow model i've trained online and run it locally with a python program I distribute. After training, I get a directory /model with two files /saved_model.pb and a folder /variables. What is the simplest way to deploy this locally? I was following here for deploying frozen models, but I can't quite read in the .pb. I downloaded saved_model.pb to my working directly and tried with tf.gfile.GFile("saved_model.pb", "rb") as f: graph_def = tf.GraphDef() graph_def.ParseFromString(f.read()) google.protobuf.message.DecodeError: Truncated message. Looking on SO here , they
订阅 google-cloud-ml-engine