No description, website, or topics provided.
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Code
Matlab
data
graph
lstm+predict
newMatlab
results
savedModels
test
LoadModelandTest.py
LoadModelandTest_lstm.py
README.md
mpccode4.m
trainUandY.py
trainUandY_lstm.py

README.md

540-Project

Data file info

File Info

File Info sineAndJump 1000,000 data points of sinusoidal function plus jumps. The frequency of the sine part is an integer from 1 to 100 Hz. The jumps are the same as "randomJump.txt". The amplitude of the jumps vary from 0.2 to 3. The amplitude of the sine wave is 3. Noise to signal ratio = 10. Essentially, I element-wise added the jump and sine references.

bigSineFile 1000,000 data points of sinusoidal functions with frequencies between 1Hz to 100Hz in steps of 1Hz. Noise added with Noise-to-Signal ratio =10. Amplitude = 1 not including noise, with noise it goes to about 1.5

randomJump 600,000 data points of random jumps to a set point. Value of set point is between -0.2 and -3 and 0.2 and 3. The jumps last for 10 to 100 time steps. No noise added

manySinesWithRef Same as bigSineFile.txt but 10,000 data points frequency ranges from 10Hz to 100Hz in steps of 10. Amplitude = 1.

manySines Same as manySinesWithRef but only contains [u y] data, not [u y ref]

testFile File for testing the model, contains 1000 data points. This is a sine wave that is perturbed a little bit.

newMPCData Not sure...