GitHub - floatingCatty/FinancialDataClustering: backup of the code and resources for Service Outsource Competition

floatingCatty / FinancialDataClustering Public

Notifications You must be signed in to change notification settings
Fork 0
Star 2

backup of the code and resources for Service Outsource Competition

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 2 Commits
.idea		.idea
Model		Model
dataProcessing		dataProcessing
utils		utils
README.txt		README.txt
loss_fn.py		loss_fn.py
test.py		test.py
train.py		train.py

Repository files navigation

——————EXPERIMENT_1————————
model is composed by encoder, K-means label, classifier
	K-means label: using mini-batch K-means clustering to labelize the data
	encoder: generalise the low_dimention representation of original data
	classifier: using the generalized data to match the K-means clustering results, to ensure the encoded vector
		maintain precise representation

the idea is mainly inspired by <Deep Clustering for Unsupervised Learning of Visual Features - Mathilde et. al.>