Search Results

You are looking at 1 - 1 of 1 items for :

  • Tornadogenesis x
  • The 1st NOAA Workshop on Leveraging AI in the Exploitation of Satellite Earth Observations & Numerical Weather Prediction x
  • Refine by Access: Content accessible to me x
Clear All
Amy McGovern
,
Ryan Lagerquist
,
David John Gagne II
,
G. Eli Jergensen
,
Kimberly L. Elmore
,
Cameron R. Homeyer
, and
Travis Smith

layer downsamples feature maps to half resolution, thus halving the spatial dimensions. Other convolutional and pooling layers perform similar operations. Feature maps from the last pooling layer are flattened into a length-6,400 vector (5 × 5 × 256 = 6,400), which is transformed by the three dense layers into vectors of length 404, then 20, and then 1. The sigmoid activation function of the final dense layer forces the output (tornadogenesis probability) to the range [0, 1]. During training

Full access