As you have seen my PR was merged and now @TensorFlow converter to #TensorFlow2.0 also supports in-place conversion for @ProjectJupyter notebooks. No need to convert to python first! Decided to record a demo, since @DynamicWebPaige had a great video on conversion yesterday! #gde

RT @RokoMijic: Since modern civilization is new, the parasites will take some time to work out how to optimally parasitize it, so if you bu….

Okay, this is officially awesome! Just use your google colab and you can generate images with stable diffusion - future is now.

RT @HuSnizhne: 23.02. Russian war preparations checklist:. 1) Gather the troops on Ukrainian border ✅.2) Occupy Belarus ✅.3) Buy internatio….

RT @hardmaru: Open-source models are going to change everything.

Pretty cool augmentation trick

And the sample code for contextual bandits on top of #tfagent #tfx

Why contextual bandits - multiple objectives and constrained optimisation among others

Anant Nawalgaria is talking how to bring research in application: Semi-supervised, Reinforcement & more #tfx #TensorFlow.

And we are back with Hacking Machine Learning meetup @hack_ai. This time as before pandemics - live in physical room with actual people. Super exciting! And the first speaker is Anant Nawalgaria from Google

RT @sumlenny: THREAD Let's start a long thread about how Russian book market prepared Russians for a full-scale war against Ukraine, NATO,….

Great story about how small steps can help solve complex problems - Testing on the Toilet #SoftwareEngineering.

Having live compose edits sounds amazing #GoogleIO #Android.

RT @benjamin_hilton: Got access to DALL-E - here, have a medieval painting of the wifi not working

Photorealistic Monocular 3D Reconstruction of Humans Wearing Clothing: "deep neural network methodology for photorealistic 3D human reconstruction given just a monocular RGB image" -

And more details on Pathways training 4/4.

Or the chain of thought 3/n

It's amazing to see such a 1-shot generalization 2/n

“PaLM: Scaling Language Modeling with Pathways”, we introduce the Pathways Language Model (PaLM), a 540-billion parameter, dense decoder-only Transformer model trained with the Pathways system. 1/n

RT @kamilkazani: Putler strategy is entirely built on assumption that the other side will play dove. If I *know* they play dove and will co….