111–120 of 218 posts
As I understand, both Jetson Nano and Google Edge TPU/Coral Dev Board would work with the same set of cameras, having the MIPI-CSI2 interface. Is it?
Many ML inference applications are using a camera, yet it's close to impossible to find something very affordable.
why would you want to train on the device itself?
Earlier quoted context omitted.
>> Nor does the TPU dev board. Yes it does require you to send your model to Google: https://coral.withgoogle.com/web-compiler/
And you have to check a box that says "I agree that my model will only be used for applications that follow Google's AI Principles." Wow.
This begs the question, what are the best price-quality ratio cameras compatible with this board? As I understand, both Jetson Nano and Google Edge TPU/Coral Dev Board would work with the same set of cameras, having the MIPI-CSI2 interface. Is it? Many ML inference applications are using a camera, yet it's close to impossible to find something very affordable.
You can also try places like Leopard Imaging or eCon systems. Even then, most of eCon's stuff is $150+ and their APIs feel a bit hacky.
This is all USB3. Unless you really need a mipi solution (you may need an adaptor board to match the), USB is fine. Even a webcam might be good enough.
I wish these tiny SBCs came with SATA or m.2 so I could hook up an SSD. I know microSD is catching up, but it's not there yet.
why would you want to train on the device itself?
In some applications you may want to run a continual learning algorithm so it will continually train on new data as well as make inferences.
the pricepoint is good, relative to the jetson tx1/2 (299-749 usd) and xavier (1099 usd). some of my notes, there is a massive heatsink on this thing, probably for both the a57 cpu and maxwell gpu, this will make your case a bit larger than say the rpi. not sure how the a57 compares to rpi’s a53, i assume both are armv8, quad-core. the inputs seem identical to rpi 3 model b+, hdmi, ethernet, 4 usb (seems 2.0), mini u…
Earlier quoted context omitted.
Because it's a M.2 E slot, which has no SATA. Maybe there are SSD drives now that can use PCI and fit in the smaller E slots? There didn't used to be. If so, great...
just no SATA? or no SATA and no NVMe?
https://www.nvidia.com/en-gb/autonomous-machines/embedded-sy...
$99 seems like a pretty good deal, am I missing anything? 4 gigs of RAM and _reasonable_ eMMC + PCI expansion could allow it to be a cheap casual use workstation, right? How does it stack up with the RK3399 in the RockPro64? I'm assuming the GPU and software support is better?
Earlier quoted context omitted.
Look at this lead pencil. There's not a single person in the world who could make this pencil. [0] [0]: Milton Friedman https://www.youtube.com/watch?v=R5Gppi-O3a8
Sorry to differ, I think this is twisted by an era where liberal market was held as god like status. Take the rubber and metal out because for now. To draw a dark line on a surface, you take the first bit of wood you find, grind it into a point and burn it. You have a pencil. I really believe that the free market centuries made people believe it was the only or most efficient way to get an object done, just like peop…
It’s not that the construction of some thing, anything, that can be used to write with is difficult. As you point out a piece of charcoal isn’t hard to make.
Milton Friedman’s point is that even something as simple and inexpensive as a pencil involves people all over the world working together to create it. Some mine graphite, some run ships to move the graphite, some manufacture paint, some grow rubber trees, etc. All of this activity, coordinated and made efficient by the market is behind even a simple thing like a pencil.