際際滷shows by User: ejlbell / http://www.slideshare.net/images/logo.gif 際際滷shows by User: ejlbell / Sat, 13 Jul 2019 20:06:04 GMT 際際滷Share feed for 際際滷shows by User: ejlbell Weak supervision - Pydata London 2019 /slideshow/weak-supervision-pydata-london-2019-155379610/155379610 weak-supervisionpydatalnd191-190713200604
An overview of weak supervision with snorkel and details of some of the issues we came across.]]>
An overview of weak supervision with snorkel and details of some of the issues we came across.]]> Sat, 13 Jul 2019 20:06:04 GMT /slideshow/weak-supervision-pydata-london-2019-155379610/155379610 ejlbell@slideshare.net(ejlbell) Weak supervision - Pydata London 2019 ejlbell An overview of weak supervision with snorkel and details of some of the issues we came across. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/weak-supervisionpydatalnd191-190713200604-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> An overview of weak supervision with snorkel and details of some of the issues we came across.
Weak supervision - Pydata London 2019 from Eddie Bell
]]> 402 1 https://cdn.slidesharecdn.com/ss_thumbnails/weak-supervisionpydatalnd191-190713200604-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Working with Fashion Models - PyDataLondon 2016 /slideshow/working-with-fashion-models-pydatalondon-2016/61790185 pydata-16-deep-160508152801
PyDataLondon 2016 presentation Fashion is a visual medium so it makes sense for our models of fashion to include visual features. In this presentation, I'll describe how we've build a general purpose visual fashion representation using CNNs. The network is multi-task (multiple labels per image), multi-image (multiple images per label) and it runs on multiple GPUs. I'll visually explore what is going on inside the black box of a neural network and discover how a fashion specific model sees the world differently from generic visual models. Lastly, I'll demonstrate a multi-modal applications of the representation learned by the model.]]>
PyDataLondon 2016 presentation Fashion is a visual medium so it makes sense for our models of fashion to include visual features. In this presentation, I'll describe how we've build a general purpose visual fashion representation using CNNs. The network is multi-task (multiple labels per image), multi-image (multiple images per label) and it runs on multiple GPUs. I'll visually explore what is going on inside the black box of a neural network and discover how a fashion specific model sees the world differently from generic visual models. Lastly, I'll demonstrate a multi-modal applications of the representation learned by the model.]]> Sun, 08 May 2016 15:28:00 GMT /slideshow/working-with-fashion-models-pydatalondon-2016/61790185 ejlbell@slideshare.net(ejlbell) Working with Fashion Models - PyDataLondon 2016 ejlbell PyDataLondon 2016 presentation Fashion is a visual medium so it makes sense for our models of fashion to include visual features. In this presentation, I'll describe how we've build a general purpose visual fashion representation using CNNs. The network is multi-task (multiple labels per image), multi-image (multiple images per label) and it runs on multiple GPUs. I'll visually explore what is going on inside the black box of a neural network and discover how a fashion specific model sees the world differently from generic visual models. Lastly, I'll demonstrate a multi-modal applications of the representation learned by the model. <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/pydata-16-deep-160508152801-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> PyDataLondon 2016 presentation Fashion is a visual medium so it makes sense for our models of fashion to include visual features. In this presentation, I&#39;ll describe how we&#39;ve build a general purpose visual fashion representation using CNNs. The network is multi-task (multiple labels per image), multi-image (multiple images per label) and it runs on multiple GPUs. I&#39;ll visually explore what is going on inside the black box of a neural network and discover how a fashion specific model sees the world differently from generic visual models. Lastly, I&#39;ll demonstrate a multi-modal applications of the representation learned by the model.
Working with Fashion Models - PyDataLondon 2016 from Eddie Bell
]]> 1596 7 https://cdn.slidesharecdn.com/ss_thumbnails/pydata-16-deep-160508152801-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Learned Representations /slideshow/learned-representations/55930843 lancaster-representations-151208120402-lva1-app6892
Presentation on learned representations for the M.Sc Lancaster data science class on features engineering]]>
Presentation on learned representations for the M.Sc Lancaster data science class on features engineering]]> Tue, 08 Dec 2015 12:04:02 GMT /slideshow/learned-representations/55930843 ejlbell@slideshare.net(ejlbell) Learned Representations ejlbell Presentation on learned representations for the M.Sc Lancaster data science class on features engineering <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/lancaster-representations-151208120402-lva1-app6892-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> Presentation on learned representations for the M.Sc Lancaster data science class on features engineering
Learned Representations from Eddie Bell
]]> 549 4 https://cdn.slidesharecdn.com/ss_thumbnails/lancaster-representations-151208120402-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 PyData London CNN Lightning Talk /slideshow/pydata-london-cnn-lightning-talk/53637428 ejlb-pydatalnd-cnn-151007092812-lva1-app6892
PyData London CNN Lightning Talk]]>
PyData London CNN Lightning Talk]]> Wed, 07 Oct 2015 09:28:12 GMT /slideshow/pydata-london-cnn-lightning-talk/53637428 ejlbell@slideshare.net(ejlbell) PyData London CNN Lightning Talk ejlbell PyData London CNN Lightning Talk <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/ejlb-pydatalnd-cnn-151007092812-lva1-app6892-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> PyData London CNN Lightning Talk
PyData London CNN Lightning Talk from Eddie Bell
]]> 1802 5 https://cdn.slidesharecdn.com/ss_thumbnails/ejlb-pydatalnd-cnn-151007092812-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 The dark art of search relevancy /slideshow/the-dark-art-of-search-relevancy/49949994 pydatalnd-search-150629093246-lva1-app6891
PyData London 2015]]>
PyData London 2015]]> Mon, 29 Jun 2015 09:32:46 GMT /slideshow/the-dark-art-of-search-relevancy/49949994 ejlbell@slideshare.net(ejlbell) The dark art of search relevancy ejlbell PyData London 2015 <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/pydatalnd-search-150629093246-lva1-app6891-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> PyData London 2015
The dark art of search relevancy from Eddie Bell
]]> 1196 2 https://cdn.slidesharecdn.com/ss_thumbnails/pydatalnd-search-150629093246-lva1-app6891-thumbnail.jpg?width=120&height=120&fit=bounds presentation Black http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 The Science of Colour (ExtractConf) /slideshow/the-science-of-colour-extractconf/49560873 extractconf-150618154153-lva1-app6892
Science of colour talk at extract]]>
Science of colour talk at extract]]> Thu, 18 Jun 2015 15:41:53 GMT /slideshow/the-science-of-colour-extractconf/49560873 ejlbell@slideshare.net(ejlbell) The Science of Colour (ExtractConf) ejlbell Science of colour talk at extract <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/extractconf-150618154153-lva1-app6892-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> Science of colour talk at extract
The Science of Colour (ExtractConf) from Eddie Bell
]]> 1032 2 https://cdn.slidesharecdn.com/ss_thumbnails/extractconf-150618154153-lva1-app6892-thumbnail.jpg?width=120&height=120&fit=bounds presentation 000000 http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 Fashion product de-duplication with image similarity and LSH /slideshow/fashion-productdeduplication/43350057 fashion-product-deduplication-150109041936-conversion-gate02
My talk from pydata London 12/2014]]>
My talk from pydata London 12/2014]]> Fri, 09 Jan 2015 04:19:35 GMT /slideshow/fashion-productdeduplication/43350057 ejlbell@slideshare.net(ejlbell) Fashion product de-duplication with image similarity and LSH ejlbell My talk from pydata London 12/2014 <img style="border:1px solid #C3E6D8;float:right;" alt="" src="https://cdn.slidesharecdn.com/ss_thumbnails/fashion-product-deduplication-150109041936-conversion-gate02-thumbnail.jpg?width=120&amp;height=120&amp;fit=bounds" /><br> My talk from pydata London 12/2014
Fashion product de-duplication with image similarity and LSH from Eddie Bell
]]> 5221 5 https://cdn.slidesharecdn.com/ss_thumbnails/fashion-product-deduplication-150109041936-conversion-gate02-thumbnail.jpg?width=120&height=120&fit=bounds presentation 000000 http://activitystrea.ms/schema/1.0/post http://activitystrea.ms/schema/1.0/posted 0 https://public.slidesharecdn.com/v2/images/profile-picture.png https://cdn.slidesharecdn.com/ss_thumbnails/weak-supervisionpydatalnd191-190713200604-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/weak-supervision-pydata-london-2019-155379610/155379610 Weak supervision - Pyd... https://cdn.slidesharecdn.com/ss_thumbnails/pydata-16-deep-160508152801-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/working-with-fashion-models-pydatalondon-2016/61790185 Working with Fashion M... https://cdn.slidesharecdn.com/ss_thumbnails/lancaster-representations-151208120402-lva1-app6892-thumbnail.jpg?width=320&height=320&fit=bounds slideshow/learned-representations/55930843 Learned Representations