ChanServ changed the topic of #mlpack to: "mlpack: a fast, flexible machine learning library :: We don't always respond instantly, but we will respond; please be patient :: Logs at http://www.mlpack.org/irc/
gtank___ has joined #mlpack
< RishabhGarg108Gi>
Hey all, I was building the project from scratch I used the simple `cmake ../` and then to speed up the process, I just did `make mlpack`. But then when I did `sudo make install`, it started to build everything including all the command line executable and tests.
< RishabhGarg108Gi>
Is this normal or I missed something ?
< RishabhGarg108Gi>
Thanks @Harsh , but I don't have any issues with the installation process. I was messing with the cmake and building mlpack in different ways to see whats happening and how cmake works.
< RishabhGarg108Gi>
I used to believe that `make` command builds the project locally in the build folder and then the `make install` simply takes those build files and puts them in the `usr/local` for using it globally. So, I didn't expect `make install` to build the remaining targets. That's what I meant to ask.
< HARSHCHAUHAN[m]>
ohh i got it 🙂
ImQ009 has joined #mlpack
< zoq>
HARSHCHAUHAN[m]: If you like 'make install' to spare out the executables, you should use the cmake configuration options like 'BUILD_CLI_EXECUTABLES=OFF'
< Antonhr>
I was trying to find out if mlpack can be cross compiled for Raspberry Pi or Jetson Nano board? Native compilation may be a little bit hard given that the compilation on a regular Laptop is fairly intensive and requires some memory.
< Antonhr>
In other words if I want to put it on a drone which has a raspberry Pi and runs Px4 flight controller which is also in C++ then mlpack looks particularly appropriate if it was supported on the Pi.
yashwants19[m] has quit [Quit: Idle for 30+ days]
Samyak has joined #mlpack
Samyak has quit [Remote host closed the connection]
< zoq>
Antonhr: Haven't tested it personally, and there is no cmake config that does the right things out of the box.
< zoq>
Ohh, right so I guess we have a cmake config that works out of the box.
< Antonhr>
ok, thank you. Since these boards run Linux and some are becoming more powerful like Jetson Nano from Nvidia and Raspberry Pi 4, they might be able to compile it natively.
< shrit[m]>
mlpack can be integrated with PX4 and RPI easily, I work on quadrotors (Gazebo simulation right now) if you have any question do not hesitate
< Antonhr>
shrit, thank you! This is encouraging. Is there documentation on how to build for the Pi ? I was trying to look on the site, but did not find how.
< shrit[m]>
Not really, the pull request is not merged yet and do not have any documentation, if you would like to build it, you can download my branch of mlpack and I will guide you during the build process
< shrit[m]>
basically you have to define the board type in cmake parameters, you have also to have the cross compiler toolchain installed
< Antonhr>
I see. I am going to keep this in mind. Will it be in the next release or at least in a different branch at some point?
< AvikantSrivastav>
Getting this error, I tried to change the g++ version but still the same error
< zoq>
AvikantSrivastav: Looks like cereal is not installed?
< AvikantSrivastav>
alright, let me try again @zoq
Antonhr has quit [Remote host closed the connection]
_slack_mlpack_40 has joined #mlpack
< shrit[m]>
o
< shrit[m]>
@Antonhr, it will need sometime, test, documentations examples, I would say we may have something in the next 6 months, but this is very optimistic.
< shrit[m]>
If you have any ideas, questions, contributions are welcome, please do not hesitate