From d99b613afba685eda0237e21a202f687b64396b6 Mon Sep 17 00:00:00 2001 From: Dawith Lim Date: Tue, 28 Oct 2025 13:19:47 -0400 Subject: [PATCH] Update README.md --- README.md | 13 ++++++++++--- 1 file changed, 10 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index d916b35..12f515c 100755 --- a/README.md +++ b/README.md @@ -21,11 +21,10 @@ The reason this is not done adaptively is because the `timeseries` blocks combin ### Encoding Required Inputs: -.mat files in the matfiles directory, .hdf5 file in hdf5 directory, or .csv files in csv directory. +.mat files in the matfiles directory (currently operational), or WORK IN PROGRESS: .hdf5 file in hdf5 directory, or .csv files in csv directory. Model inputs (x): -* A matrix containing 130 feature vectors corresponding to frequency, each 26 tokens long, corresponding to number of measurements. -* (Work in progress) 3x26 matrix containing backscatter brightness, normalized standard deviation and backscatter brightness. +* A matrix containing 130 channels corresponding to frequency, each 35 tokens long, corresponding to number of measurements, and 3 channels each 35-tokenslong containing backscatter brightness, normalized standard deviation and foreground pixel count. The 130x35 matrix and the three 1x35 vectors are individually regularized and merged into one 133x35 matrix before being fed into the neural net. Encoder output (y): * Two 1xN one-hot encoded classification vectors that predicts sample type and treatment applied. @@ -33,6 +32,14 @@ Encoder output (y): To run encoding, use `python train_encoder.py`. This is subject to change later when the decoder is actually operational and there is a need to distinguish the training code file names. ### Decoding +Required Inputs: +.mat files in the matfiles directory (currently operational), or WORK IN PROGRESS: .hdf5 file in hdf5 directory, or .csv files in csv directory. + +Model inputs (y): +* Two 1xN one-hot encoded classification vectors that predicts sample type and treatment applied. + +Decoder output (x'): +* One 133x35 matrix that can be split into 130x35 spectra, 1x35 backscatter brightness, normalized standard deviation, and foreground pixel count. To run decoding, use `python decodertest.py`. This is subject to change later as this is only a "test" version of the decoder training workflow.