google-cloud-automl

来源： https://stackoverflow.com/questions/60326319/coreml-not-returning-proper-identifier-tags-using-automl-vision

问题 Now i can use autoML node.js client library to train the model on google-cloud-automl. Q: How can i programmatically get the model id when finished training the model?. Goal: I will use that id to deploy the model without web interface. Tried: At first, i thought it is in the response when training the model (operation.name). But the operation.name showed projects/${projectId}/locations/${location}/operations/${operationId} , which is not include model id. So i have no idea how to

问题 Now i can use autoML node.js client library to train the model on google-cloud-automl. Q: How can i programmatically get the model id when finished training the model?. Goal: I will use that id to deploy the model without web interface. Tried: At first, i thought it is in the response when training the model (operation.name). But the operation.name showed projects/${projectId}/locations/${location}/operations/${operationId} , which is not include model id. So i have no idea how to

问题 I've been using AutoML Vision Edge for some image classification tasks with great results when exporting the models in TFLite format. However, I just tried exporting the saved_model.pb file and running it with Tensorflow 2.0 and seem to be running into some issues. Code snippet: import numpy as np import tensorflow as tf import cv2 from tensorflow import keras my_model = tf.keras.models.load_model('saved_model') print(my_model) print(my_model.summary()) 'saved_model' is the directory

问题 I am using automl from google using a custom dataset. The dataset consists of images collected by me. However, manually labelling the image take some time, so I would like to enlarge the dataset by image augmentation, such as rotation and blurring. Does automl perform augmentation behind the screen automatically? 回答1: AutoML does few types of data augmentation. This is implementation detail and may change in the future without notice, basic augmentations that are used are: random resizes /

问题 So basically I am trying to load a model trained using AutoML's Tables with tensorflow. Once a model is trained, it can be exported to Google Storage going to TEST & USE -> USER YOUR MODEL -> Container. After downloading it locally, the model files look like this in my file system: model/ ├── assets │ ├── 14_vocab │ ├── 15_vocab │ ... │ ├── 7_vocab │ └── 8_vocab ├── saved_model.pb └── variables ├── variables.data-00000-of-00001 └── variables.index I've tried to loading the model in several

问题 I am currently trying to build an app that uses google's autoML feature. I have trained a model and published it on google firebase and have integrated the necessary code into my app following the documentation: https://firebase.google.com/docs/ml-kit/ios/label-images-with-automl I am using a remote model instead of making it local. However when I try running the code, then choose an image in the simulator, an empty list of predictions is output in the console. I have also turned on the

问题 after training an image classification model using AutoML Vision, the "Evaluate" tab gives access to model performance, specifically to the confusion matrix. My dataset has over 100 labels, but the confusion matrix is only displayed as a 10x10 matrix. My questions: does the confusion matrix only display the worst performances (above a certain threshold for example) or is this limited to a 10x10 representation regardless of performance? Is there a way to access the entire confusion matrix?

问题 I loaded a dataset into google automl, using the UI. I got the message that I have enough labeled text and can start training, however when I click on Start Training, I get the error Exception while handling your request: Request contains an invalid argument. When reporting refer to this issue by its tracking code tc_698293 As I am using the UI, I don't know what the arguments of the request are. Any help is greatly appreciated. Thanks. 回答1: It is required to have at least 2 examples in all

问题 I get different results when using a model to get image annotation predictions from web UI and from API. Specifically, using the web UI I actually get predictions, but using the API I get nothing - just empty output. It's this one that gives nothing using the API: https://cloud.google.com/vision/automl/docs/predict#automl-nl-example-cli Specifically, the return value is {} - an empty JS object. So, the call goes through just fine, there's just no output. Any hints as to how to debug the issue