- Non-reactivity - body sensation
- Non-reactivity - working with sound
- Non-reactivity - thinking
- Non-reactivity - emotion
- The judging mind
- Mental noise as addiction
- External cues as mindfulness reminders
- Beginners mind
- One step at a time
- Grasping and aversion
- Before you speak, let your words pass through three gates: Is it true? Is it necessary? Is it kind? — Rumi
Jeremy H.: There's dropout, weight decay, batch normalization and data augmentation. You pretty much always want batch normalization. Data augmentation we will see in a moment. So then it's really between dropout vs. weight decay.
I have no idea. I don't think I've seen anybody to find a compelling study on how to combine those two things. Can you always use one instead of the other? Why? Why not? I don't think anybody has figured that out.
I think in practice, it seems that you generally want a bit of both. You pretty much always want some weight decay. But you often also want a bit of dropout. But honestly, I don't know why.
I've not seen anybody really explain why or how to decide. So this is one of these things you have to try out and kind of get a feel for what tends to work for your kinds of problems. I think the defaults that we provide in the Python library should work pretty well in most situations but yeah, definitely play around with it.
Jeremy H.: ❗ Remember that [L2 regularization
class AdaptiveEmbedding(nn.Module): | |
def __init__(self, n_token, d_embed, d_proj, cutoffs, div_val=1, | |
sample_softmax=False): | |
super(AdaptiveEmbedding, self).__init__() | |
self.n_token = n_token | |
self.d_embed = d_embed | |
self.cutoffs = cutoffs + [n_token] | |
self.div_val = div_val |
If you plan to quickly put together a simple web app or website with React.JS.
YMMV depending on what you’re doing, but the following is a good bet if you want to make the project accessible to other developer, and it doesn't need to quickly scale.
Use React.JS with TypeScript.
Create React App now makes it dead easy. Just run this command:
interface Array<T> {
concat(...items: Array<T[] | T>): T[];
reduce<H>(
callback: (state: H, element: H, index: number, array: T[]) => H,
firstState?: H): H;
···
}
This is a third-party/un-official implementation of the paper Drop an Octave: Reducing Spatial Redundancy in Convolutional Neural Networks with Octave Convolution (OctConv).
- Python 3
- Tested with Python 3.6
Here are the best startup tools of 2019 that will help you build out your startup business as quickly, cheaply, and efficiently as possible.
This is a curated list of tools for everything from productivity to web hosting to development tools to designing. Most of these tools are either free or have limited free option that is enough for startups. We love all the free services out there, but it would be good to keep it on topic. It's a bit of a grey line at times so this is a bit opinionated; feel free to suggest and contribute in this list.
- GitHub — Unlimited public repositories and unlimited private repositories (up to 3 collaborators).
- GitLab — Unlimited public and private Git repos with unlimited collaborators.
- BitBucket — Unlimited public and private repos (Git and Mercurial) for up to 5 users with Pipelines for CI/CD.
- Visual Studio — Unlimited private repos (Git a
The theme for this year is "Building a more helpful Google for everyone".
- Google Lens
- Duplex on the web
- Google Assistant - 100 GB DL model to 0.5 GB
- voice is faster than typing (tapping) your phone
- AI and bias, fair for everyone
- Zebra model + TCAV
- Data Privacy & Security
- Privacy