Hguimaraes / gtzan.keras

[REPO] Music Genre classification on GTZAN dataset using CNNs
MIT License
198 stars 57 forks source link

Where is the number 300 comming from? #25

Closed galloj closed 2 years ago

galloj commented 3 years ago

In file https://github.com/Hguimaraes/gtzan.keras/blob/master/nbs/1.1-custom_cnn_2d.ipynb theres number 300 used in majority voting, can you please explain where does this number comes from? It would be really nice to know, because after changing amount of genres and songs this part of script stops working.

willnufe commented 3 years ago

My English is not very good. First, you should know the y_test.shape is (11700, 10). After the operation of np.argmax(y_test, axis=1), the shape of y_test is turned into (11700,). Each sound was divided into 39 segments in our custom function, the test dataset has 300 songs. so 11700 = 39 * 300.