Just some stuff I did with Ap data from 1991 to 2016. First of all, it is clear that ap data is not continuous, and appears in seperate bins (as expected). Highest values around 400. x axis is time (in hours since 1 1 1991). The Ap index increases as frequency decreases (sorted version of […]
MTF results- this problem 1D CNN: Input example: Outputs: Training Loss figure: Clearly better than the MTF method in my opinion- more decisive and more accurate.
Just a summary of some cool things I have done in Leeds so far. I helped somebody with a parameter problem by creating a map of the results of a machine learning clustering algorithm in parameter space- that was fun. I found that negative Chi values worked well for the OPTICS algorithm (though I am […]
This post is password protected.
I just finished applying an LSTM to Space weather data (Sunspot number over the last few hundred years.) It is a work in progress, but I am happy that I have at least managed to start to do what I was intending to. The x-axis is in terms of months (like I said, it is […]
This week, I have been doing some oceanography training. I will present the map below and expect it to be self-explanatory in terms of the overall tone. It is clear that accurate measurements show that average sea level rises by one centimetre a year, which is slightly concerning. At one station (Newlyn, UK), the following […]
Happy new year! I completed some more work on the ARIMA dataset before Christmas that I wanted to share. I firstly applied the same method as I used in previous analyses, plotting the RMS error over the full forecast time to see how the different methods react to longer forecast times. Using the full 1991-2016 […]
I recently posted a blog post to IAGA, and I thought it would make sense to post it here as well. Got some wonderful feedback on the context image (featured), though it was the NOAA who provided it! Aurora events on 3rd and 4th November 2021: A summary People up and down the United […]
Hi. Currently on week 3 of training. What have I done so far? Well, I just did a classification on some satellite data, and it looked pretty cool, so that’s something. I started with a map of Europe (thank you Google Earth Pro!) and classified it into 5 classes (deep water, water, forest, […]
I have done an RMS graph in a similar fashion to the graph I did last week, and I have come up with the following plot. It looks a lot more promising for the ARIMA forecast compared to the previous plot. I included 3 datasets this time (from 2014, 2015 and 2016). It is clear […]