########################################### # Dataset Usage ########################################### Step 1: ======= Download the datataset from http://aseados.ucd.ie/datasets/EMSCA Step 2: ======= Check the MD5 hash value with the following command. md5sum em-dataset.h5.gz The value should be the following. 876564812f06c6689563339f2b784650 em-dataset.h5.gz Step 3: ======= Decompress the file using the following command. gzip -d em-dataset.h5.gz Step 4: ======= Calculate the MD5 hash value for the uncompressed HDF5 file. md5sum em-dataset.h5 The value should be the following. b998495b45e7aea27e1912ea060d404d em-dataset.h5 Step 5: ======= Viewing dataset and loading data into NumPy arrays using the provided python3 reading-hd5-file.py Now you can use these EM trace data in NumPy arrays for training and testing ML models. Appendix: ========= The dataset in HDF5 format was created using from the raw EM trace files as follows. # creating the HDF5 file using the raw (.cfile) EM trace files. python3 creating-hd5-file.py # compressing to with level 6 gzip -6 -c em-dataset.h5 > em-dataset.h5.gz