Skip to content
Snippets Groups Projects
Select Git revision
  • Chutchins_Branch_Feature1
  • Elliot_Reborn
  • Jzhou_Amazing
  • master default protected
  • spencer_branch
5 results
You can move around the graph by using the arrow keys.
Created with Raphaël 2.2.029Jun15329May1910754330Apr25232221201918161514131292131Mar30272524232019181312119854227Feb2625242322211912-More CleaningChutchins_Branc…Chutchins_Branch_Feature1Merge remote-tracking branch 'origin/Chutchins_Branch_Feature1' into Chutchins_Branch_Feature1-Generally Sorted out the repository a bit more.mastermaster-Reverting Changes in iridis_setup_python.sh-Added a Requirements file-Change File Paths in two models-Tidying up tflite conversion code-Commented some file paths outAdded file paths for Linux to GenerateMels_TUT2017.py-I can generate a tflite model! you just need to build a new environment with TF 2.5 because nothing else works apart from that... HORAH! Such Tensorflow. Much Unreliable.-Given up with tflite model-Updated Model 13 with FP CNN changes-Generally Sorted out the repository a bit more.-Reverting Changes in iridis_setup_python.sh-Added a Requirements file-Change File Paths in two models-Tidying up tflite conversion code-Commented some file paths outAdded file paths for Linux to GenerateMels_TUT2017.py-I can generate a tflite model! you just need to build a new environment with TF 2.5 because nothing else works apart from that... HORAH! Such Tensorflow. Much Unreliable.-Given up with tflite modelMerge remote-tracking branch 'origin/Chutchins_Branch_Feature1' into Chutchins_Branch_Feature1- Added some code to convert model to a tflite model (Not done yet)-Updated Model 13 with FP CNN changes-Start of benchmarking stuff-Updated Learning Rate and Learning Rate Decay in Model_CH_12.py-Updated Learning Rate and Learning Rate Decay-Changed Iridis Script-Changed File pAths back to Iridis-Added batch norm to one of the layers in model 12, seemed to help with performance-Changed FP kernel size to 3,3-Added a run_on_iridis_gpu.sh file. For, surprisingly... running on gpu nodes ;)-Change to Iridis file paths-Binarised the Model, Added Jing's BOP optimiser and took some parameters from Model 11-Created Model 12 which implements validation and test loss into the tensorboard metricsCreated a new CRNN model, implemented with some of Cagdas' suggestions.-Added CNN Model, Full Precision with custom Adam Learning rate and DecayBOP gamma=1e-3, gamma=1e-6Jzhou_AmazingJzhou_AmazingBOP gamma=1e-6, gamma=1e-3BOP gamma=1e-6, gamma=1e-3