HackerTrans
TopNewTrendsCommentsPastAskShowJobs

codeinassembly

no profile record

Submissions

Show HN: Efficient `Torch.cdist` Using Triton

github.com
1 points·by codeinassembly·el año pasado·0 comments

Show HN: Connecting MLflow Hosted Models and Label Studio for Auto Labelling

5 points·by codeinassembly·hace 2 años·0 comments

Show HN: BoilerBot – Amazon Alexa Prize Challenge 2 TaskBot Code, Purdue Team

github.com
4 points·by codeinassembly·hace 2 años·0 comments

[untitled]

1 points·by codeinassembly·hace 3 años·0 comments

Explaining Large Language Models by Breaking Them

dagshub.com
3 points·by codeinassembly·hace 3 años·1 comments

Deep Learning Papers Are Kinda Bullsh-T

dagshub.com
96 points·by codeinassembly·hace 4 años·64 comments

Unraveling the Deep Learning Reproducibility Crisis

dagshub.com
11 points·by codeinassembly·hace 4 años·0 comments

A TensorFlow Re-Implementation of CheXNet – Localization of Thoracic Diseases

2 points·by codeinassembly·hace 4 años·0 comments

comments

codeinassembly
·hace 3 años·discuss
And other fun details
codeinassembly
·hace 4 años·discuss
Neat!
codeinassembly
·hace 4 años·discuss
It is available via ONNX, which can convert to tensorflow weights. From there, it's possible to perform post-training quantization and it should finally be available for use with Coral. However, there's a good chance operations aren't yet supported by their chip, and accuracy will surely take a hit.