-
Hi,
I have run the pretraining.py data on domain specific text.
I gave 9 lakh sentences text file, batch size 32, learning rate 2e-5, num of train steps 10000.
masked lm accuracy - 69%
Aft…
-
-
**Residual Network가 왜 잘 되는지 해석해보기**
-Deeper is better? -> No, there is degradation problem.
-ResNet은 이 문제를 뒷 단으로 미룬다.
-ResNet은 그냥 쓰면 다 잘 되었지만 '왜' 잘되는 지에 대한 해석은 없었다.
-왜 잘되나?
: ResNet is an…
-
### Is there an existing issue for this?
- [X] I have searched the existing issues
### Feature Description
Deep Belief Networks (DBNs) are a type of generative graphical model that consist of multi…
-
Hello, I’m Cem Gunes
I have questions about labeling my custom models. I have read the paper, checked other closed issues; however, I could not apply a clear pipeline to prepare my models as parts …
-
## Keyword: sgd
### Doubly Stochastic Models: Learning with Unbiased Label Noises and Inference Stability
- **Authors:** Authors: Haoyi Xiong, Xuhong Li, Boyang Yu, Zhanxing Zhu, Dongrui Wu, Dejin…
-
I know. A stupid and naive question. But I am a beginner and I am struggling to find a major use case for my workflow so maybe if the advantages can be spelled out, that would be even better!
Also…
-
Part of _succeeding_ in our [`Mission`]() is making sure all the people on our team `feel loved` and `feel` like they are doing great work solving real human challenges with technology and making peop…
-
# Project Context
There are certainly many topics that are good for a data scientist to know and practice! I've tried my best so far to research and develop what I would consider an outline of the…
-
Subject: A huge thanks for your help!
Hey [Colleague's Name],
I wanted to take a moment to express my sincere gratitude for your help during my onboarding process. Your guidance and in-depth knowledge…