Model Codes can be found in the FinalModelCodes
folder. To run the training and generate the same results, all the codecells can run in order, after creating a shortcut to our dataset in your drive and passing the path to the dataset.
Project Report can be found Project Report.pdf
Weights can be found in the weights
folder.
Dataset can be found here and here.
EE603-Machine-Learning-Signal-Processing
│ Decoding.ipynb
│ FeatureComparision.ipynb
│ LICENSE
│ Project Report.pdf
│ README.md
│
├───DatasetPipeline
│ DatasetGeneration_FINAL.ipynb
│ Generate_Test_Data.ipynb
│
├───FinalModelCodes
│ CNN_FINAL.ipynb
│ DNN_FINAL.ipynb
│ GMM_final.ipynb
│ LogReg_final.ipynb
│ RNN_FINAL.ipynb
│ TemplateMatching_final.ipynb
│
├───OldTestCodes
│ ClassifierOldMFCCVariant.ipynb
│ DatasetGeneration.ipynb
│ GMM.ipynb
│ LogReg.ipynb
│
└───weights
│ gmm_weights.npy
│ logRegWeights.zip
│
├───CNN_FINAL_MFCC_CORRECT
│ │ keras_metadata.pb
│ │ saved_model.pb
│ │
│ └───variables
│ variables.data-00000-of-00001
│ variables.index
│
├───DNN_FINAL_MFCC_CORRECT
│ │ keras_metadata.pb
│ │ saved_model.pb
│ │
│ └───variables
│ variables.data-00000-of-00001
│ variables.index
│
└───RNN_FINAL_MFCC_CORRECT
│ keras_metadata.pb
│ saved_model.pb
│
└───variables
variables.data-00000-of-00001
variables.index