Differences
This shows you the differences between two versions of the page.
| Next revision | Previous revision | ||
|
forecasting:meeting_minutes_december_6_2016 [2016/12/07 02:52] jobatake created |
forecasting:meeting_minutes_december_6_2016 [2021/09/19 21:59] (current) |
||
|---|---|---|---|
| Line 3: | Line 3: | ||
| ===== Updates ===== | ===== Updates ===== | ||
| - | * | + | * Final Presentation was good |
| + | * Working on Paper | ||
| + | * Help Brianne with poster | ||
| ===== Meeting with Dr. Kuh ===== | ===== Meeting with Dr. Kuh ===== | ||
| ===== Expectations ===== | ===== Expectations ===== | ||
| - | * Check Accuracy | + | * Batch |
| - | * RMSE: 200 | + | * Test and Train with Seasons |
| - | * Persistence: Use last data point as estimate | + | * Online |
| - | * Current time = prediction | + | * LMS: First approach |
| - | * 1 tap = 1 | + | * Recursive?: Second approach |
| - | * No bias | + | * Predict solar energy |
| - | * No normalization | + | * Non-Linear? |
| - | * Create estimate: All data and guess mean | + | * Support vector machines |
| - | * Max solar irradiance is 1000 W | + | * Generate data (Gaussian) |
| - | * Check with Travis's Report | + | * Tap delay line |
| - | * Test many data sets | + | * Uncorrelated linear gaussian becomes correlated gaussian |
| - | * Test RSME should be higher than train RMSE | + | * UC Irvine Data Repository |
| - | * Varies on the number of taps | + | * Feel comfortable with regression |
| + | * Law of Large Numbers | ||
| + | * Average = Expected Value | ||
| + | * Find variance | ||
| + | * Optimal weight = (R-1)P = 1/1.64 | ||
| + | * P = 1/m(XD) | ||