Assembly Line Activity Recognition: Difference between revisions

The network was tested on a separate part of the dataset, that had not been presented to the network. The best performing model achieved an accuracy of 0.91832, a precision of 0.91333, a recall value of 0.91832 and finally an f1-score of 0.91521. These metrics indicate a very high performance of the network for this particular task;  the following table presents the complete training, validation and testing results.  

Finally the inference demo below shows a gif of the assembly process under inference; it displays the top prediction for the sample as well as the associated softmax value. This inference process has a threshold of 0.5 meaning that if none of the softmax values achieve a 50% value, then no activity is recognized.