I am trying out Azure Cognitive Services OCR to scan in an identity document. It works fairly well but I was wondering if it is possible to train the OCR engine or somehow link it to a learning service to improve character recognition ?
I don't think that you can train Azure OCR, but there is one new Azure service called Form Recognizer which gives better results than the previous OCR service and also you can train it on custom data.
Related
We have 60+ azure analysis services in our subscription so how can we get size of azure analysis size? So we want to automate this and publish in front end reports where users can see information.
It is difficult to get size of each cube in each azure analysis service by logging to each azure analysis service using SSMS.
Going with azure metrics memory option is also not right and accurate option.
Following below blog, but it is not allowed me run script in powershell ISE, got some error.
How we get analysis services database size in azure analysis services Tabular model
Is there any option to get all azure analysis services size using single script or any REST API ?
Thanks for your help.
Regards,
Brahma
I have a requirement wherein I need to fetch historical data from a third party system which is exposed as a web service and train the model on that data.
I am able to achieve the above requirement by using "Execute Python Script" node and invoking the web service using python.
The main problem arises when I need to fetch data from the third party system every time the Azure ML web service is invoked, since the data in the third party system keeps on changing hence my Azure ML model should be trained for new data always.
I have gone through the link (https://learn.microsoft.com/en-us/azure/machine-learning/machine-learning-retrain-a-classic-web-service) but I am not sure how we can do this for my requirement as for me the new historical data set should be obtained every time the Azure ML web service is invoked.
Please suggest.
Thanks.
I recommend that you:
look into the new Azure Machine Learning Service. Azure ML Studio (classic) is quite limited in what you can do, and
consider creating a historical training set stored in Azure blob storage for the purposes of training, so that you only need to fetch from the 3rd party system when you have a trained model and would like to score the new records. To do so, check out this high-level guidance on how to use Azure Data Factory to create datasets for Azure Machine Learning
I am using the Google Translate neural network (amazing improvement) via the Google Cloud Translation API in SDL Trados to process technical translations.
Of course it needs heavy post-editing, mostly terminology and sometimes style. I would really like if the neural network could learn from this post editing - but there seems to be no way to do feed my edits back.
It is possible when using the web interface manually (translate.google.com).
The (years unupdated) Google Translator Toolkit allowed to used a shared public TM, but that is now obsolete with the neural network.
Can I somehow feed translations back to Google Cloud Translation API to train it?
Their FAQ states this:
"Does Google use my data for training purposes?
No, Google does not use the content you translate to train and improve our machine translation engine. In order to improve the quality of machine translation, Google needs parallel text - the content along with the human translation of that content."
As you pointed out, in the documentation regarding confidentiality, it is highlighted that Google does not use the data for training purposes as a background/transparent process, due to the following reasons:
Confidentiality: for confidentiality reasons, the content inputted to the Translation API will not be used for training the model.
Non-feasibility: the Neural Network model behind Translation API would require the non-translated content plus the translated version suggested by the user in order add some training to the model; so it is not possible to train the model with just the non-translated text.
Moreover, there is currently not the possibility to suggest translations to the API in order to train the model in a more custom way.
As a side note, you might be interested in keeping an eye on AutoML, the new Google Cloud Platform's product that is currently still in alpha, but to which you can request access by filling in the form in the main page. It will allow the creation of custom Machine Learning models without requiring the dedication and expertise that other more complex products such as ML Engine require. The first product of the AutoML family to be launched will be AutoML Vision, but it is possible that similar products will appear for some of the other ML-related APIs in the Platform, such as the Translation API, which is the one you are interested in.
Also feel free to visit the Google Cloud Big Data and Machine Learning Blog from time to time in order to keep updated in the latest news in this field. If you are interested in AutoML, its release and presentation will probably have an article in the blog too.
So as a summary: no, currently you cannot feed suggested translations back to the Translation API, but in the future you might be able to do so, or at least have your own custom models.
I want to create an application where in I capture images of people within my family and detect who it is. can I use vision API to create cloud database to store different pictures of each family members labelled with their names so that when I pull a request from the API it scans the images from the database and detects which family member it is rather than just detecting the faces in it. can I train it to do so???
It is possible to train the classifier. More details can be found in this link: How to train and classify images using Google Cloud Machine Learning and Cloud Dataflow
I created an experiment within Azure ML Studio and published as a web service. I need the experiment to run nightly or possible several times a day. I currently have azure mobile services and azure web jobs as part of the application and need to create an endpoint to retrieve data from the published web service. Obviously, the whole point is to make sure I have updated data.
I see answers like use azure data factory but I need specifics as in how to actually set up the scheduler.
I explain my dilemma further # https://social.msdn.microsoft.com/Forums/en-US/e7126c6e-b43e-474a-b461-191f0e27eb74/scheduling-a-machine-learning-experiment-and-publishing-nightly?forum=AzureDataFactory
Thanks.
Can you clarify what you mean by "experiment to run nightly"?
When you publish the experiment as a web service, it should give you and api key and the endpoint to consume the service. From that point on you should be able to call this api with the key, and it would return the result processing it tru the model you've initially trained. So all you have to do is to do the call from your web/mobile/desktop etc application in the desired times.
If the issue is to retrain the data model nightly, to improve the prediction, then this is a different process. That was only available tru the UI only, now you can achieve this programmatically by using the retraining api.
Kindly find the usage of this here.
Hope this helps!
Mert