In this post, I would like to share my practice with Facebook’s new Detectron2 package on macOS without GPU support for street view panoptic segmentation. If you want to create the following video by yourself, this post is all you need. This demo video clip is from my car’s dashcam footages from Preston, Melbourne. I used the PyTorch and Detectron2 to create this video with segmentation masks.
This post is following upgrade with respect to the early post How to Build an Artificial Intelligent System (I) The last one is focused on introducing the six phases of the building an intelligent system, and explaining the details of the Problem Assesment phase.
In the following content, I will address the rest phases and key steps during the building process. Readers can download the keynotes here: Building an Intelligent System with Machine Learning.
Phase 1: Problem assessment – Determine the problem’s characteristics.
What is an intelligent system?
The process of building Intelligent knowledge-based system has been called knowledge engineering since the 80s. It usually contains six phases: 1. Problem assessment; 2. Data and knowledge acquisition; 3. Development of a prototype system; 4. Development of a complete system; 5. Evaluation and revision of the system; 6. Integration and maintenance of the system .
Aurecon’s experts, across Cape Town, Melbourne and Auckland offices, have been teamed up to develop and test approaches that capture and validate new and existing measurements of the metropolitan road network. Due to the confidentiality, we reduced the resolutions of the aerial images and only opened limited results on the public domain at https://roadsfromabove.netlify.com/. Thanks to Greg More, the design of this website got the best feedback from the workshop (Visualization for AI Explainability) of the IEEE VIS 2018 conference in Berlin, Germany
Visualization for AI Explainability: Projections and Dimensionality Reduction. The goal of this workshop is to initiate a call for “explainable” that explain how AI techniques work using visualizations. We believe the VIS community can leverage their expertise in creating visual narratives to bring new insight into the often obfuscated complexity of AI systems.
When I am using TensorFlow on my MacBook Air, I always get annoyed by the warnings comes from nowhere, so I followed the documentation below to build TensorFlow sources into a TensorFlow binary and installed it successfully. In theory, this will make the TF running faster on my machine.
Here is the document:
If you are a Mac user, you could download the TF binary from here:
Then, you could use conda to initialize an environment with Python=3.6 and install TF by typing:
sudo pip install tensorflow-1.8.0-py2-none-any.whl
This is the 3rd post about my implementation of TensorFlow Apps on my Android Phone.
This time I fixed one small bug in the app of “TF Detect” so the object tracking function could work. The project is compiled by “cmake“ with NDK Archives in this version. You can download the new “apk files here: Tensorflow_Demo_Debug.apk.
“Once the app is installed it can be started via the “TF Classify”, “TF Detect”, “TF Stylize”, and “TF Speech” icons, which have the orange TensorFlow logo as their icon.
Python is a great general-purpose programming language on its own, but with the help of a few popular libraries (Numpy, SciPy, Matplotlib, TensorFlow) it becomes a powerful environment for scientific computing and data analysis.
I start sharing my notebooks on learning Python and TensorFlow, here is my GitHub repository: Data_Science_Python.
Great news! PyTorch now is supporting Windows!
If you have a PC with suitable Nvidia graphics card and installed CUDA 9.0 and Anaconda, type the following commands;
conda install pytorch cuda90 -c pytorch
pip3 install torchvision
It is about 500 MB, so be patient!
Underline is the old post.
PyTorch is a deep learning framework that puts Python first. Currently, it only supports MacOS or Linux.
But, can we use it on WIN10 without changing the system/computer?
Yes, we can.
This post is a follow-up to my last post: A Taste of TensorFlow on My Android Phone.
You could download my compiled Apk file here and install on your android device (>Android 7.0) (22-May-2018).
Update: 25-May-2018, fixed the bug of object tracking function in TF Detect. PS: While running the activities, pressing the volume keys on your device will
toggle debug visualizations on/off, rendering additional info to the screen that
may be useful for development purposes.
- tensorflow_demo_debug.apk (My compiled version: 4 Apps ~ 109.8 MB );
- TfLiteCameraDemo.apk (official version: 1 App ~ 20.05 MB).
Feel free to let me know if there are any bugs 🙂
In the last post, I just tried the TensorFlow for object classification (TF Classify). This time I installed all four demos of the TensorFlow Mobile for Android according to this tutorial: TensorFlow Lite Demo for Android. They are awesome 😛