Hacker News Reader

Comment by eanzenberg | Hacker News Reader

Is this a subtle ad for clarifai?

For the record, I've worked on building a classifier for nudity detection and not much get's better than starting with a pre-trained model and fine-tuning to your specific rules. I wouldn't be surprised if this is the future of deep learning: Start with the closest pre-trained model that has trained for months using custom clustered hardware and fine-tune further on commodity hardware.

View on HN

Have you looked at the Yahoo Open NSFW model? I ported it to Algorithmia if you want to try it out to compare:

https://algorithmia.com/algorithms/spullara/YahooOpenNSFW

Ya I have, it is the 1 that does best so far besides a custom fine-tuned model. It beats google, msft and clairifai.

That's not the future, it's what is standard in ML application engineering. For example, a good portion of classification systems use models based on ImageNet, VGG, Ngram or AlexNet and then fine tune the last (or last few) layers.

Transfer learning is an old and well known technique and is used ubiquitously. [1]

[1]http://cs231n.github.io/transfer-learning/

That's a very good business model as well to sell these pre-trained models. Like a market where you choose what you need.

Or, better, you submit a pre-tagged sample set and it tells you which models match and how well.

That's what Algorithmia is trying to do.

[1]https://algorithmia.com/tags/machine-learning

I think this is true not just of images but also for text with things like word2vec etc

Reply on news.ycombinator.com