Protesters out in force to demonstrate against CANSEC arms show. I asked McGuinty's office for his comments about the barring of journalists who might write critical pieces about his speech/ the govt defence policy. He isn't commenting. ottawacitizen.com/public-ser… via @ottawacitizen

The UK once had some of the best railways, public housing, a solid welfare state that took care of people and a more caring/loving society. Then Thatcher became Prime Minister.

🚨BREAKING UPDATE🚨 Tyre is burning again tonight. Only hours after the earlier blitz on the city, Israeli strikes resumed after midnight, setting residential neighbourhoods ablaze while civilians tried to shelter for the night. These images are from Tyre now. Apartment buildings on fire. Streets lit by explosions. Civil defence crews rushing into the flames as the bombing continues around them. Israel is lashing out across south Lebanon with escalating fury, not only to terrorise civilians, but also out of frustration at its growing inability to stop Hezbollah’s FPV drone attacks. Netanyahu appears to believe that flattening neighbourhoods & intensifying collective punishment will force the resistance into submission. It won’t.

Another participant in the Canadian Global Sumud Flotilla, Daniela Maria Bonamico, has arrived in Toronto after Israel illegally detained and tortured activists. Look at the crowd at Pearson Airport in Toronto.

We are releasing Carbon: a crazy fast DNA model Carbon is 275x faster than the next best model. So fast you can process the whole human genome on a single GPU in <2 days. Here are the tricks we used: When modelling DNA sequences a lot of the performance comes down to tokenizing the sequences in a smart way. BPE tokenizer struggle because there are no whitespaces and character (called base in DNA) level tokenizers waste a lot of compute on too many tokens. Carbon is built with a unique tokenizer: we split sequences in chunks of 6 bases, but during both training and inference we can work with single base resolution. That's similar to having word tokens but resolving them at the character level. All possible thanks to the DNA tokens unique structure. The architecture combined with the tokenizer makes the model 275x faster than the previous SoTA (Evo2) at this size. We built an interactive demo so you can explore how the model can generate DNA sequences, investigate the structure of genes, predict the effect of mutations, generate and fold proteins and even reconstruct parts of the tree of life. huggingface.co/spaces/Huggin…