Congcong Li

[Congcong Li @lufficc]

A maskrcnn-benchmark-like SSD implementation, support customizing every component! And EfficientNet-B3 backbone is support now! Details can be found on GitHub: https://t.co/VfnlPCKfcI . @PyTorch @fvsmassa

[Congcong Li @lufficc]

@jaywcjlove 这是哪个项目的？

[Congcong Li @lufficc]

@Blankwonder OK多谢🫡

[Congcong Li @lufficc]

Start a focused work session with todo planning and a lightweight pomodoro timer in your browser.

https://t.co/6Rgy7yxWBd

[Congcong Li @lufficc]

还记得你何时加入 Twitter 吗？我知道！#我的Twitter周年纪念日

[Francisco Massa @fvsmassa]

I'm very happy to announce the 0.3 release of torchvision. It brings several new features, including custom C++/CUDA ops, pre-trained Mask R-CNN models and much more! Check it out at https://t.co/izgRz0UQVV

Plus, training Mask R-CNN is even faster than in maskrcnn-benchmark!

[Congcong Li @lufficc]

😃

[François Fleuret @francoisfleuret]

My deep-learning course materials are now updated to PyTorch 1.0.0.

https://t.co/a3AaGEG3Bq

Close to 900 slides! Comments are welcome.

#deeplearning #idiap #epfl #pytorch

[Congcong Li @lufficc]

#PyTorch PyTorch 1.0 稳定版放出了，不得不说自从 0.4 之后 PyTorch变的好用很多。那我基于 PyTorch 1.0 实现的 SSD 也公开吧 😍https://t.co/VfnlPCKfcI

[PyTorch @PyTorch]

[v1.0] : JIT Compiler, Faster Distributed, C++ Frontend, CUDA10
Read more about the changes at https://t.co/Xfrad06NZO

As always, get the install commands on https://t.co/DeaBDSRxs8

[ONE PIECE スタッフ【公式】/ Official @Eiichiro_Staff]

ついに除幕！

ルフィ像、ここに立つ！

[Andrej Karpathy @karpathy]

ResNet-50 on ImageNet now (allegedly) down to 224sec (3.7min) https://t.co/3Z77Edfj0u using 2176 V100s. Increasing batch size schedule, LARS, 5 epoch LR warmup, synch BN without mov avg. (mixed) fp16 training. "2D-Torus" all-reduce on NCCL2, with NVLink2 & 2 IB EDR interconnect

[Congcong Li @lufficc]

PyTorch 真香😂😂😂!

[Congcong Li @lufficc]

Great post !
https://t.co/bF9h1LTNkP

[Congcong Li @lufficc]

Mojave 又是一堆 Bug，果然抢先更新就要做小白鼠🙃🙃🙃😑