The attached ipynb file contains code for electrical fault prediction on the GPVS-Faults Experimental Dataset. The data consists of two power modes: Limited Power Mode and High Power Mode. Four models were trained, two for each power mode. The models trained were an LSTM model and a CNN-LSTM hybrid model. The highest accuracy achieved is 85% in Limited Power Mode by the LSTM model, and 83% in High Power Mode by the Hybrid Model