Note on Automatic 3D Instance Segmentation Pipeline In this note I try to summarize several recent works on Automatic 3D Instance Segmentation, with most direct application to saturated reconstruction of neural morphology in an imaging volume (mostly scanning Electral Microscopy, but seems it can be generalized into other imaging modality), which is one of the most important method of high-throughput connectomics1.
How to automatically analyze behavior video? DeepLabCut is a powerful tool to rapidly1 train a neural network (based on ResNet) to track keypoints on movement videos, esp. those of moving human or animals. Thus this is a game changing tool for all kind of behavior quantification for neuroscience and psychology researchers (can be applied to nearly any behavioral science topic, e.g. motor learning, motor control, facial expression, social interaction…). The workflow is relatively simple and it scarcely takes time after the network have been trained, and the video analysis can be done automatically. Because of this it’s really favorable to the reserchers doing long term ecological video recording.
Overview Objective: Build a online blog. Most preferably Posts can be written in Markdown, which is the format of most of my notes. Math formula (in Latex syntax) and equation rendering are Solution
Just like written languages, different programming languages have different innate styles, which suit different needs. Like Matlab love matrix, love script; python favors tensor, favors subfunction, favors objective-oriented programming; Mathematica are muchly functional; R are better at datatable demonstration.
Vector Image The key thing is, the image is not represented by an array of pixel values, it’s an file recording the points and lines. Thus you can visualize the image at different resolution and different scale.
This is a technical demo of how to translate a full TF biggan to pytorch. https://github.com/ajbrock/BigGAN-PyTorch/blob/master/TFHub/converter.py Note
NLTK Old API to Stanford NLP set CLASSPATH=E:\DL_Projects\NLP\stanford-segmenter-4.0.0\stanford-segmenter.jar set STANFORD_SEGMENTER_PATH="E:\DL_Projects\NLP\stanford-segmenter-4.0.0" A pure python demo of this API. os.environ['CLASSPATH'] = r"E:\DL_Projects\NLP\stanford-segmenter-4.0.0\stanford-segmenter.jar" # this is used to run the jave code os.environ["STANFORD_SEGMENTER"] = r"E:\DL_Projects\NLP\stanford-segmenter-4.0.0" # this is used to search for dictionaries segmenter = StanfordSegmenter() segmenter.default_config('zh') res = segmenter.segment(u"北海已成为中国对外开放中升起的一颗明星") print(res) # segmenter = StanfordSegmenter( # path_to_sihan_corpora_dict=r"E:\DL_Projects\NLP\stanford-segmenter-4.0.0\data", # path_to_model=r"E:\DL_Projects\NLP\stanford-segmenter-4.0.0\data\pku.gz", # path_to_dict=r"E:\DL_Projects\NLP\stanford-segmenter-4.0.0\data\dict-chris6.ser.gz" ) This is a really well documented demo of old API
basicConfig() to configure the root logger works only if the root logger has not been configured before. Basically, this function can only be called once. https://realpython.com/python-logging/ https://docs.python-guide.org/writing/logging/
Motivation for PyTest Each test file can share local variables in that script. In PyCahrm you can configure test setting in Python integrated tools>testing PyTest Fixture Some components are shared among multiple tests (like a neural network), we can use a fixture to initialize it for all the tests.
Optimizer Development for Online Neural Activity Maximization Motivation Background Here, I state a few technical constraints on Optimizer Development, and our way to tackle it. Zeroth Order (Derivative Free): Limited evaluation number: Neural variability: For sensory cortices, same stimuli can elicit different response in different presentation. High dimensionality of stimuli space: For visual world and auditory world, the dimensionality for natural inputs are high. Geometry and topology of the stimuli space: Methodology Many of the existing work on online neural activity optimization are using Genetic Algorithm, which is a classic evolutionary algorithm. As all evolution algorithm has these components