Hey Tanmay! I really loves this learning series! Is there a 5th lesson that follows on from this one?
@sohannikumbh4802
3 жыл бұрын
You are dropping amazing content on your channel...... KEEP IT UP 👍👍
@leonine5216
3 жыл бұрын
Thank you Tanmay really enjoying this series 👍
@anandiborade6349
3 жыл бұрын
Hey Tanmay, when are you going to release the next video in this series??
@MokshithSriVishnu
3 жыл бұрын
Thanks everyone for subscribing my channel
@rne1223
3 жыл бұрын
Great series. I hope you can finish it.
@sakshidubey4149
3 жыл бұрын
Hey Tanmay can I get the link of the 5 th video of this series ? Thank you
@boris554
3 жыл бұрын
Hey, can you say something about that white hat jr scam
@rne1223
3 жыл бұрын
At 14:29 I see the "one shot" function but it only last 1 frame. Was that on purpose?
@gauravmehta9450
2 жыл бұрын
But distances in higher dimensions will be larger right? how can we compare the distances in a higher dimension and match them with the distance in the lower dimension?
@tanmayteaches
2 жыл бұрын
Not necessarily - that really depends on the data. Just being in a higher dimension space doesn’t mean two points must be farther from each other than two points in a lower dimension space. In fact, in some cases, data may converge to the opposite. For example, most neural networks end up embedding data in very high dimension spaces very very close to each other, since the fact that there are so many dimensions means that it can manipulate that space in much more complex ways, and therefore it’s easier to keep things closer to each other. However, it is true that it can be impossible to preserve every exact distance between every point when moving from a higher dimensional space to a lower dimensional one. This is why we use gradient descent, it can help us find an approximation that preserves as much information about the distance between points as possible.
@lukes5631
5 ай бұрын
@@tanmayteaches Does this mean that using small numbers that are very to close to one another across many dimensions would result in a better storage of data, i.e. more dimensions = more space to store data? And if so, would this be more efficient to store data this way?
@pramodkhade2554
3 жыл бұрын
Do you understand math model and algorithm behind this ?
@tanmayteaches
3 жыл бұрын
Hi! That’s exactly what I’m explaining in these first few videos. Please take a look and let me know what you think - I’d be happy to answer any questions you may have.
@CrusadeVoyager
3 жыл бұрын
Hi Tanmay , Nice talk, can we have the link to your code in github?
@tanmayteaches
3 жыл бұрын
Hi! Here's a link to the code: gist.github.com/tanmayb123/9d7b0bdcf2663d77f048e7c48ac44182
Пікірлер: 25