3 ms·
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-
by eanzenberg 9y ago
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.
- strzzz 9y agoThat's a very good business model as well to sell these pre-trained models. Like a market where you choose what you need.
- fancy_pantser 9y agoOr, better, you submit a pre-tagged sample set and it tells you which models match and how well.
- AndrewKemendo 9y agoThat's what Algorithmia is trying to do. [1]https://algorithmia.com/tags/machine-learning https://algorithmia.com/tags/machine-learning
- AndrewKemendo 9y agoThat'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/ http://cs231n.github.io/transfer-learning/
- spullara 9y agoHave 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 https://algorithmia.com/algorithms/spullara/YahooOpenNSFW
- eanzenberg 9y agoYa I have, it is the 1 that does best so far besides a custom fine-tuned model. It beats google, msft and clairifai.
- alexcnwy 9y agoI think this is true not just of images but also for text with things like word2vec etc