Does not work with kaggle data.

[ignvinay]

Hi micah5, Its great work for developing classification based on audio. I tried your code, but seem to not work properly. I tried training and testing with kaggle data, which you can also try https://www.kaggle.com/mmoreaux/audio-cats-and-dogs/version/5#cats_dogs.zip

It seem to be dropping one label and also misclassifying all the time. I increased epochs and got close to 90% accuracy, but still then seem to do wrongly. Even test data given is training data , that also seem to not work.

Sample output :

(tf_gpu) d:\Vinay\Audio_classification\pyAudioClassification\example>python test.py Using TensorFlow backend. 2019-04-15 15:53:10.947087: I tensorflow/core/platform/cpu_feature_guard.cc:141] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX AVX2 2019-04-15 15:53:11.788533: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1432] Found device 0 with properties: name: GeForce GTX 1050 major: 6 minor: 1 memoryClockRate(GHz): 1.493 pciBusID: 0000:01:00.0 totalMemory: 4.00GiB freeMemory: 3.30GiB 2019-04-15 15:53:11.792768: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1511] Adding visible gpu devices: 0 2019-04-15 15:53:12.162764: I tensorflow/core/common_runtime/gpu/gpu_device.cc:982] Device interconnect StreamExecutor with strength 1 edge matrix: 2019-04-15 15:53:12.172438: I tensorflow/core/common_runtime/gpu/gpu_device.cc:988] 0 2019-04-15 15:53:12.173703: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1001] 0: N 2019-04-15 15:53:12.175952: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1115] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 3018 MB memory) -> physical GPU (device: 0, name: GeForce GTX 1050, pci bus id: 0000:01:00.0, compute capability: 6.1) Lets predict the file data/cat/cat_1.wav

1. dog 0.0% (index 0) [[6.274987e-21]]