exp-time-series-tools / ms-cnn Goto Github PK
View Code? Open in Web Editor NEWMulti-Scale Convolutional Neural Network for Time Series Classification (MCNN)
Home Page: https://www.cse.wustl.edu/~z.cui/projects/mcnn/
Multi-Scale Convolutional Neural Network for Time Series Classification (MCNN)
Home Page: https://www.cse.wustl.edu/~z.cui/projects/mcnn/
Source: https://www.cse.wustl.edu/~z.cui/projects/mcnn/ ######################################################### # Python packages ## ######################################################### 0. numpy (version I use: 1.10.4) 1. Theano (version I use: 0.7.0) Useful link for installing theano: http://deeplearning.net/software/theano/install.html ######################################################### # How To Run ## ######################################################### mcnn.py trains on one of the ucr dataset, Trace, with multiscale convolutional neural network (MSCNN). It can run on both CPU and GPU. 1. run 'python mcnn.py' ## This command trains a MSCNN with default parameters. ## You should be able to get zero error on train set, ## validation set as well as test set. 2. run "THEANO_FLAGS='blas.ldflags=-lblas -lgfortran,mode=FAST_RUN, cuda.root=/usr/local/cuda,device=gpu,floatX=float32, lib.cnmem=1' python mcnn.py" ## This command trains exactly the same MSCNN as the first one. ## However, You can enjoy 10x speedup if you have GPU with ## CUDA installed using this command. 3. run 'python mcnn.py -h' for more information. ######################################################### # Standard Output ## ######################################################### increase factor is 29 , ori len 275 train size 2320 ,valid size 580 test size 2900 batch size 232 n_train_batches is 10 data dim is 371 --------------------------- building the model... training... ...epoch 1, valid err: 0.75000 | test err: 0.43000 | train err 0.73621, cost 2.2027 ...epoch 2, valid err: 0.75000 | test err: 0.43000 | train err 0.65603, cost 1.4417 ...epoch 3, valid err: 0.75000 | test err: 0.43000 | train err 0.54526, cost 1.0825 ...epoch 4, valid err: 0.75000 | test err: 0.43000 | train err 0.48319, cost 0.9583 ######################################################### Please contact me if you have any question.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.