NVIDIA Jetson Nano 2GB is the Best Way to Learn AI

Video Statistics and Information

Video
Captions Word Cloud
Reddit Comments
Captions
in march 2019 nvidia released jetson nana this is really interesting board for the robotics and for the eye because it's not expensive and it's super powerful this year in may they released even more powerful board jets on xavier nx but it's a little bit pricey and now they released one more board one more developer kit jetsan nana 2 gigabyte this is cheap and quite powerful computer for the ai this should be really interesting computer to learn the ai and the most interesting thing that if you develop your project with this computer in mind afterwards you can replace it with a more powerful browser and it still should work because they really similar in terms of the software and in terms of the hardware let's open the box and look what is inside this is where you get your nvidia products here it is that's another 2 gigabyte developer kit perfect this is two boxes side by side this is original jetson nana and this is jetsan nana two gigabyte this is ninety nine dollars and this is fifty nine dollars so this one is way cheaper and as you can see immediately original jetson number has four usb ports all these ports is usb 3.0 and over here there is only three usb ports one usb port is 3.1 and to other usb 2.0 enough this boring stuff let's open the box unboxing time and let's take this jetson out it has usbc for the power this is a connector for the camera the same as on a raspberry pi and actually you can use the camera from the raspberry pi with this one the standard gpio connector everything looks very similar to the original jetson nano over here there is a connector for fan if you would like to put one but i think it should work perfectly fine without any additional active cooling and this is original jetson nana so you can check them side by side looks pretty much similar it looks like this is usb 3.1 and these two usbs are 2.0 there is also pins over here and there is different stuff on these pins like leds serial connector power button and stuff like this this is jetsang xavier nx this is jetsan nana and jetsam nana 2 gigabyte all of them together now let's power up this one and see how it works have already prepared the micro sd card it goes here at the back this is a keyboard this is a wi-fi it was included and hdmi from the screen now it's only left the power usbc give me the power cool it works apparently this is ubuntu now let's install the software for the real sense camera because this is quite important for the robotics for this you can follow either the instructions on the jetson hacks or on the intel itself intel real sense so i'm going to follow the instructions on the intel now let's connect our camera this is my real sense depth camera and you need usb 3 for it and now we need to run real sense viewer it found the camera d435 let's turn on this stereo module and so our real sense camera works perfectly fine there is no errors [Music] yeah and so this is the second real sense camera i think it's t265 and so this is a tracking camera real sense viewer try 3d and it works perfectly fine i connected jets and nana with remote with this usb cable to the laptop it's actually a funny and easy way to connect the jets and nana to your laptop and one of the most interesting things that you can connect to just some nana using arduino program this is so funny and this is really in the diy spirit so i use the arduino serial monitor and you see it's possible to connect it i also connected this camera let's see all the fun stuff that we're going to do with it so you can remotely run the jupiter notebook like this one this is actually from deep learning institute so this is one of this nice tool with which you can learn deep learning quite easily and interactive yeah so you run all of these you run all of this blah blah blah blah blah blah blah and afterwards you get this one we have camera here we have the live stream from this camera and we can add different photos to our data set and afterwards we can train our model on these photos something like this for example uh we can add some sub ups some down and now we can train our model and now we have the life over here and there is some up and some down so it sinks now that it's some down ha ha ha this is cool so there is a lot of stuff with which you can play by the way this green board is a board which my friend is developing for me and with this board you can control actuators using the jetsa nano or raspberry pi or jetson xavier and all this is possible because they all have the same gpio connector and even the holes they're at the same position so it's like really perfect i will make video about this board later because uh we still developing it it's uh we don't yet have the firmware for it over here on the table i summarized some stuff which i personally think is interesting about this jetson nana 2 gigabyte affordable jetson nana this jetsanana is good for the education and for the robotics let's start with the education many my viewers asking me how to learn robotics and my answer is that i think the best way to learn something is to try to do it so even if you don't know robotics just try try something try something simple afterwards a little bit more complicated etc etc etc but i personally think that in order to learn something you need to do something with your hands like uh really because when you're doing something yourself you find out the problems and you're trying to solve these problems etc etc and this is exactly the purpose of the jetsana you try the deep learning by doing this on the real hardware and so first of all there is a free nvidia courses for this after you do the courses from the deep learning institute of the nvidia and you submit your project and if your project is good enough they will give you the certificate and this would be perfect for example when you're looking for a job in this domain as i said i really like that it runs on the real hardware and not just on cloud it's even powerful for model training of course you wouldn't be able to train super complicated models but for simple models it will work and in most of the cases it's more powerful than the competitors and now about robotics most interesting part so first of all you can run the pipe bullet like this you can do your robotics in theory in simulation you can run ros i did not do this i don't know how to do this but apparently it's possible i need to learn this you can connect your real sense camera depth camera it works perfectly fine no errors it's compatible with most raspberry pi heads it has a gpio connector so like this you can easily control hardware so this tool is kind of the same it has low power consumption this is not really important for my projects because usually my motors they consume way more than just 10 watts but for me i see one big drawback that it has only one usb 3.1 and so i cannot connect my two real sense camera at the same time but what i really like that uh jets and nana two gigabytes so this cheap version is also hardware and software compatible with the jetson xavier nx at least mostly compatible so this means that you can start with jetson nana 2 gigabyte this affordable version you can develop your project and when you're happy with this project and when you out grow jets another two gigabyte you can buy the more powerful browser jetson xavier nx and theoretically it should immediately work in your project don't forget to put the like to this video i'm serious if you would like that i do some project with machine learning write this in the comment thank you for watching my video till the end huge big thank you to my patrons here's their names this is the best people who help me with this channel as usual stay safe good luck with your projects and see you next time you
Info
Channel: Skyentific
Views: 40,690
Rating: undefined out of 5
Keywords: robotics, robot, skyentific, gadget, cool gadget, robotic arm, Arduino, DIY, jetsonnano, Nvidia, Jetson nano, how to learn ai
Id: NZGH9iu3O1E
Channel Id: undefined
Length: 10min 35sec (635 seconds)
Published: Mon Oct 05 2020
Related Videos
Note
Please note that this website is currently a work in progress! Lots of interesting data and statistics to come.