AidLearning for Linux-Python-AI code building on Android. AidLearning framework builds a linux environment supporting GUI, deep studying and visual programming on Android devices..Now Beef up is helping TensorflowLite GPUDelegate by the use of python.
Intro
AidLearning is an impressive cell AI building platform, which is helping just about all frameworks and tool for deep studying neural neighborhood building.
- It has built a complete Linux OS supporting GUI desktop on Android, built-in the most popular deep studying framework Caffe / mxnet / keras / Python / tensorflow / ncnn / opencv/pytorch…
- In-built Visual AI building IDE, built-in freshest programming equipment similar to VSCode and Jupyter, supporting touch-and-drop ui design, supporting dynamic debugging and dealing of code. Beef up the development of AI methods in Python on cell and pad, and improve the package your Python provide code into app (APK) to liberate.
- One click on on arrange is supported. You most simple want to arrange a 10m app to robotically boot to complete the arrange.
- At the moment, it is been online in number one app software amenities, with more than 1.5 million downloads and visits, and a large number of AI examples and tutorials are built-in. There are also a large number of tutorials on the Internet to facilitate your studying and building.
- Finally, on type 0.87,Beef up supplies CPU acceleration and GPU acceleration, and tflite is built into type 0.87. tflite GPU module can completely liberate the GPU potency of mobile phone, improve CPU + GPU mode, and boost up at the an identical time! tflite.NNModel (model_ Path, inshape, outshape, 4,0) # – 4 represents 4 CPU threads, 0 represents GPU, – 1 represents CPU, and 1 represents the number of nnapi threads. I set 4 threads on lend a hand. You’ll be able to flexibly set the number of threads and use GPU + CPU mode
Choices
Innovation
- The most efficient Linux environment on android terminal (mobile phone), the only Linux os that is helping GUI desktop without vnc operating on the android.
- It is the most straightforward simulator that is helping AI building environment, built-in deep studying framework with the sphere’s freshest best 7, and a large number of deep studying models, examples and building components
- The only simulator that is helping Python Visual building and debugging is helping touch and drag interface design to strengthen your building efficiency
- It is helping the development of apps that can run on cellphones with Python, and is helping the direct compilation of Python code to generate deployable APK information
- One click on on arrange, without any dependence, you most simple want to arrange a 10m boot app in your mobile phone, and you are able to robotically complete the arrange of all environments.
- Move platform building, improve cloud desktop (cell desktop and computer desktop are the equivalent), can’t most simple run on mobile phone or tablet or other embedded motherboard, however moreover may also be directly accessed and complicated on the computer facet in step with Web.
- It is helping openblas, multi thread and multi process, runs simply without jamming, and gives whole play to the computing power of ARM CPU and GPU
Versatility
- Beef up Tensorflow, Caffe, mxnet, keras, python, ncnn, opencv, Scipy
- Is helping Python 2.7 / Python 3.7.3.
- Buildin AidCode visual programming IDE, it moreover is helping Jupyter pocket guide and Microsoft’s vscode programming building software.
- In-built complete and native move platform desktop, no want to arrange third-party VNC improve, improve the equivalent desktop of computer and mobile phone
- It is helping now not most simple mobile phone/pad, however moreover industrial arm board
- The complicated program may also be deployed on each and every mobile phone and computer
- It is helping 99.5% of the cellphones available on the market, and has tested a whole number of 64 bit cellphones similar to Huawei, vivo, oppo, Samsung and Xiaomi
- Beef up Linux native xfce4 desktop, do not want to arrange VNC and other software
- Beef up the development of pyqt5, pyGame, vortex, SDL, and so forth
Coverage
- Beef up virtualizes a closed area on the mobile phone. It does now not need root and won’t harm the content material subject matter of your mobile phone.
- Isn’t going to obtain your own privacy, all permissions may also be set by the use of yourselves
Easy to use
- One click on on arrange, robotically download the latest dependency package, robotically configure the environment required for AI building, and reduce the threshold of AI building
- In-built a large number of AI components, models, examples, tutorials, reduce the threshold of AI building, you are able to now not understand AI algorithm, alternatively you are able to use this platform to increase AI methods.
- The built-in sensor regulate package can merely regulate various sensors on the mobile phone: sound, gyroscope, position, camera, and so forth
- One mobile phone, two methods, Android and Linux co-exist, no restart, loose switching; recreational, building, studying 3 now not unsuitable
- It is helping the automatic synchronization of mobile phone building and computer building code, is helping interface touch and drag design, and robotically generates interface code
- One click on on compilation and liberate of AI enabled apps
- Extensible improve Java, C + +, transfer… And other languages
improve peripherals
- The built-in sensor regulate package can merely regulate various sensors on the mobile phone: sound, gyroscope, position, camera, and so forth
- The usage of OTG USB can improve the extension of peripherals and regulate aduino, which may also be programmed in Python
- The usage of OTG USB can also improve peripheral storage tool be told and write operations
- It can be used since the operating device of suave robot
Construction
Aidlearning framework may also be divided into two parts: Linux simulator and AI programming platform.
Linux simulator consists of terminal and desktop. The former builds a complete Linux simulator in step with Android underlying Linux kernel and busybox command package, and you are able to arrange any dependency package you need with apt command; the latter builds a graphical operating desktop in step with Web, which you are able to regulate all the device with touch in your mobile phone. At the an identical time, the desktop is helping cloud desktop, which you are able to merely get right to use through a internet web site on the computer.
AI programming platform is composed of deep studying framework and python visual programming framework (Python IDE). The former incorporates just about all the popular deep studying framework, which is in control of the loading of models and the scheduling of calculation graphs, and incorporates the memory allocation and op implementation of each calculation. After that, a python visual rapid building platform is built, which can’t most simple run and debug Python code online, however moreover improve touch-pull interface design, and generate the whole executable program and output APK record.
Rapid get began
Buildin Apparatus
Enlargement
FeedBack
License
Thanks
flay、gondon、willam、gugu、yoline777、qidiso、yuge…
- VTE (libvte): Terminal emulator widget for GTK+, principally used in gnome-terminal. Source, Open Issues, and All (including closed) issues.
- iTerm 2: OS X terminal software. Source, Issues and Documentation (which comprises iTerm2 proprietary escape codes).
- Konsole: KDE terminal software. Source, particularly tests, Bugs and Wishes.
- hterm: JavaScript terminal implementation from Chromium. Source, along with tests, and Google group.
- xterm: The grandfather of terminal emulators. Source.
- Connectbot: Android SSH client. Source
- Android Terminal Emulator: Android terminal app which Termux terminal coping with is in step with. Inactive. Source.
- Termux: Android terminal and Linux environment – app repository. Source.
- remi:Python REMote Interface library. Platform unbiased. In about 100 Kbytes, absolute best for your diet.[Source] (https://github.com/dddomodossola/remi).
- [Caffe]https://github.com/BVLC/caffe
- [Tensorflow]https://github.com/tensorflow/tensorflow
- [Mxnet]https://github.com/apache/incubator-mxnet
- [Keras]https://github.com/keras-team/keras
- [ncnn]https://github.com/Tencent/ncnn
- [pytorch]https://github.com/pytorch/pytorch
- [opencv]https://github.com/opencv/opencv
