Backend updates, increasing number of calls and image upload via URL
It has been a busy month in our Prague office. Two new team membersjoined us and we are at full throttle to get rid of the BETA label. We experienced a couple of service outages connected to scaling and increasing number of API calls.
Send image as URL
This month we added a possibility to send API calls with an link to the image. You can simply replace a “image” for “url” and send a link as a value. This comes handy for the users who store their images in cloud services like AWS and Google Cloud.
You can also use both cl-api.vize.ai and cl-api.vize.it domains now.
API scaling and outage
Our API is facing much more calls than last month. We are working on the smooth handling of scaling which is not an easy task. We did some architectural changes and this allowed us to hold more models and smaller response time. We apologies for any inconvenience and recommend staying in touch with us in the case of sensitive applications.
We now run multiple trainings in parallel. This will reduce the time needed for your task to be trained. We keep our fine-tuned models and so it still takes about 2–3 hours to train different variants of models but there is not a delay caused by task queue anymore.
Continuous model improvement
If you are interested in automated continuous model improvement stay tuned and shoot us an email to get more info (become a private tester). We are going to bring amazing features to upgrade your models and datasets.
Start recognition with Vize.ai
Vize.ai provides accurate image classification for specific recognition tasks. You can train custom models in our user interface and test them in Preview. Contact us on moc.r1566318920alimi1566318920x@sel1566318920as1566318920 for more info or read more about:
- How to prepare image dataset?
- How to define an image recognition task?
- How to train image classifier in 5 minutes?
- How does machine learning work? Like a brain!
This article was originally published by David Rajnoch.