For our course projects, we started off with Google's AIY Vision kit. I am looking into how to use output for next level processing. We are all newbies to ML
I am testing image classification using sources from Github AIY Projects on my mac.
python image_classification.py -i images/cat01.jpeg
FileNotFoundError: [Errno 2] No such file or directory: '/opt/aiy/models/mobilenet_v1_160res_0.5_imagenet_labels.txt'
I would like to test this to see output format that I want to use for next steps.
Where can I get the labels txt file?
I found a good Coral link that has TPU models and labels. Is there a way to use this instead?
couldn't get it to work on my mac but tried to create a dockerfile for some of the AIY example codes to run on a docker host within the mac.
The missing piece to overcome the error you listed above for me was to :
apt-get install -y aiy-models after adding the Debian repo source. Not sure how to get around that on the mac.
Related
I am trying to do a simple simulation in Cooja (literally the example on the Contiki website here).
However I have an error message linked to my #include "net/rime.h", Cooja outputs the following message while compiling:
fatal error: net/rime.h: No such file or directory
I thought that it came with Contiki when I've installed contiki-ng following the procedure on the official website. My quick fix was to download the file from the github repository and store it in my active directory. However I find it quite ugly, is there a way to install this library ? Because I haven't found a way yet.
Thanks a lot.
It's a question close to this one.
Which kind of method are you using to start Cooja? Directly on the SO? Over Docker?
I recommend you de Docker way, and remember to apply the command
git submodule update --init --recursive
after cloning the git repository.
Also, if you are using Docker, check if the binding of the Contiki folders between de Host and the Container are correctly made.
The official website makes it pretty clear that there is no support for kenlm in Windows. There is a Windows tag at the github repository but it seems to be maintained by few random contributors then and there.
How to set up kenlm for Windows then?
The new DeepSpeech PlayBook also includes instructions for setting up a Docker image and running training from within a Docker container. If you have Docker on Windows, this might be another solution.
The information for building a new Scorer is still in a PR, but may also be useful.
The solution is to use Ubuntu in Windows through Windows Subsystem for Linux
Get WSL for Windows
From your ubuntu bash navigate to the folder where you want to do the setup. You can access the Windows file system from the /mnt/c/ folder, which you can find at the root directory.
From there simply follow the official instructions, that is clone the git repo, and run cmake .. & make -j2 in order to build the project (after first making the necessary installations in your Ubuntu system).
Obviously, you must train the models or scorers using the Linux bash. You can also use these models from Windows using the kenlm python library.
E.g.
The two steps to build a scorer for the deepspeech-model as described here should be executed from your Ubuntu system. But after you have the scorer you should be able to run the command
deepspeech --model deepspeech-0.9.3-models.pbmm --scorer kenlm.scorer --audio audio.wav
from Windows. However, once you have WSL there's no need to do this work from Windows. Things will work nicely #your Ubuntu system.
I've faced the same problem and solved it by building kenlm wheel from Cygwin terminal as home page advices (pip wheel pypi-kenlm).
I've also uploaded wheel to pypi called kenlm-cygwin, but it's only python3.7.
Can I get a tip for installing on rasp buster? Im hung up on the install directions to check the status of the rabbitMQ server. Traceback of bash console:
(volttron) pi#raspberry:~/Desktop/volttron $ echo 'export RABBITMQ_HOME=$HOME/rabbitmq_server/rabbitmq_server-3.7.7'|sudo tee --append ~/.bashrc
export RABBITMQ_HOME=$HOME/rabbitmq_server/rabbitmq_server-3.7.7
(volttron) pi#raspberry:~/Desktop/volttron $ source ~/.bashrc
pi#raspberry:~/Desktop/volttron $ RABBITMQ_HOME/sbin/rabbitmqctl status
bash: RABBITMQ_HOME/sbin/rabbitmqctl: No such file or directory
There are a few tracebacks earlier on the installation...
If it makes a difference or not here is the entire bash console process. The git gist link I just created the name install.py even though its just bash commands copied pasted per install directions...
`pi#raspberry:~/Desktop $ git clone https://github.com/VOLTTRON/volttron --branch releases/7.x`
It looks like there are a couple of different issues going on here:
The issue you quote above (RABBITMQ_HOME/sbin/rabbitmqctl: No such file or directory) is that your shell isn't finding the rabbitmqctl command. It looks like you added the RABBITMQ_HOME environment variable to your .bashrc, but used the string RABBITMQ_HOME instead of the variable expansion $RABBITMQ_HOME when you tried to run the command. Try running it as $RABBITMQ_HOME/sbin/rabbitmqctl status instead.
The rabbitmqctl status command will check the status of the rabbitmq application, but I don't think you've done anything to start it yet (that happens when you bootstrap the platform and/or start the platform configured to use the RMQ broker)
I think that the traces earlier in the installation process are problematic (appears to be the same error hit two different ways), but you just haven't run into them yet. I haven't seen any issues building gevent on the RPi 4 with buster (though it is pretty slow), but the ctypes error makes me wonder if there's an issue with the underlying c library it is trying to build on top of. I did notice that you're getting amd64 erlang packages, are you running Raspbian on an x86 processor? (if so this isn't a permutation we've tried and you may be hitting some package compatibility edge case we haven't seen)
One thing to try is to manually install cython into your virtualenvironment and then try running the bootstrap script again with the virtualenvironment activated. You could also try and pip install gevent==20.6.1 directly in that virtualenvironment (this is what the bootstrap script was doing at the failure point). VOLTTRON depends on gevent, so if that isn't installing the platform won't be able to run.
I am trying to use Flink local on Linux and Windows, for my bachelor
thesis. I have found these steps for local setup:
https://ci.apache.org/projects/flink/flink-docs-release-1.1/quickstart/setup_quickstart.html#start-a-local-flink-cluster
When I try this I got only errors like this:
-bash: bin/start-local.sh: No such file or directory
When I go to the directory of the start-local.sh file then I got
/flink-1.1.2/flink-dist/src/main/flink-bin/conf/flink-conf.yaml: No such file or directory
Same problem with Windows.
What do I have to change so that it works?
It seems that you have downloaded the sources. It is necessary to download one of binaries from here: https://flink.apache.org/downloads.html#binaries. Then, follow the given instructions for local setup.
Of course if you want to build Flink from sources, use this guide: https://github.com/apache/flink#building-apache-flink-from-source.
The dronekit Getting Started page suggests installing WinPython to use dronekit-Python on Windows because it includes the dependencies. I already have a working Python installation and I prefer not to risk messing it up with WinPython. What are the dependencies I need to install?
As of DKPY 2.0 this is outdated. Also, I might move to making a MavProxy module depending on whether or not the unpaid devs decide to stay when 3DR stops funding Dronekit
I've written a procedure to help with this problem which I've pasted. 3DR claims they're going to fix it, but in the mean time I hope this will help.
This setup is for Windows 64-bit systems only, although similar procedures will work with 32-bit.
Install MAVProxy and run it once before reaching step 5.
Install Notepad++.
Install Python v2.7.
Inside the Python folder, run WinPython Control Panel and select Advanced->Register Python.
Inside the same folder, run WinPython Command Prompt and input the following four commands:
• pip uninstall python-dateutil
• pip install droneapi
• pip install console
• echo module load droneapi.module.api >> %HOMEPATH%\AppData\Local\MAVProxy\mavinit.scr
Install WX Python. It should be the 64-bit Python 2.7 version.
Download and install OpenCV 2.4.11 to any folder
• Copy/paste the file cv2.pyd from OpenCV\build\python\2.7\x64\ to \python-2.7.6.amd64\Lib\site-packages.
Steps 8 through 11 apply to SITL only
Follow the online documentation for setting up Cygwin for SITL in Windows
Go to C:\cygwin\home\Your Username\ardupilot\Tools\autotest\
Open sim_vehicle.sh in Notepad++
• Change line 429 from…
cygstart -w "/cygdrive/c/Program Files (x86)/MAVProxy/mavproxy.exe" $options --cmd="$extra_cmd" $*
to...
cygstart -w "/cygdrive/c/Users/YOUR USERNAME HERE/Desktop/WinPython-64bit-2.7.6.4/python-2.7.6.amd64/Dronekit/Scripts/mavproxy.py" $options --cmd="$extra_cmd" $*
Note: This location changes depending on where you installed WinPython. For me, it was the desktop.
Start simulations as you would normally for SITL. To run Python scripts during the simulations, use the command
• api start Path to script\script_name
To use the code to connect to an actual copter, open WinPython Command Prompt
• Navigate to the folder which contains the scripts you wish to test
• Type mavproxy.py --master=”com##”,57600
• Run your script by typing into the MAVProxy terminal
o api start script_name