AnshThakur

All_conv_net_IS2018

# All Conv-Net for Bird Activity Detection Significance of Learned Pooling Bird activity detection (BAD) deals with the task of predicting the presence or absence of bird vocalizations in a given audio recording. In this work, we propose an all-convolutional neural network (all-conv net) for bird activity detection. All the layers of this network including pooling and dense layers are implemented using convolution operations. The pooling operation implemented by convolution is termed as learned pooling. This learned pooling takes into account the inter featuremap correlations which are ignored in traditional max-pooling. This helps in learning a pooling function which aggregates the complementary information in various feature maps, leading to better bird activity detection. Experimental observations confirm this hypothesis. The performance of the proposed all-conv net is evaluated on the BAD Challenge 2017 dataset. The proposed all-conv net achieves state-of-art performance with a simple architecture and does not employ any data pre-processing or data augmentation techniques. This work is accepted for publication in INTERSPEECH 2018. feature_extract.py is used to extract melspectrogram features. all_convnet_BAD.py is the all convolutional model architecture all_convnet_BAD_maxpool_variant.py is the maxpool variant of the all_conv_net model get_activation_map.py is used to analyse any layer activations in the model To know more about BAD 2017 challenge and to download data, follow this link http://machine-listening.eecs.qmul.ac.uk/bird-audio-detection-challenge/

CCSE_ICASSP

COMPRESSED CONVEX SPECTRAL EMBEDDING FOR BIRD SPECIES CLASSIFICATION

CDT-TimeSeries

continual_EHR_JACOB_A

Continual Learning of Electronic Health Records (EHR).

Contrastive_loss_keras

Face verification using contrastive loss

DAA_IMK

DE

Directional embedding

FL4HeterogenousEHRs

Demo for "Federated Learning For Heterogeneous Electronic Health Records Utilising Augmented Temporal Graph Attention Networks"

NC_Scores_Demo

Quantum-Encoding

Quantum Encoding For Healthcare

AnshThakur/CDT-TimeSeries

AnshThakur/Contrastive_loss_keras

Face verification using contrastive loss

AnshThakur/FL4HeterogenousEHRs

Demo for "Federated Learning For Heterogeneous Electronic Health Records Utilising Augmented Temporal Graph Attention Networks"

AnshThakur/All_conv_net_IS2018

# All Conv-Net for Bird Activity Detection Significance of Learned Pooling Bird activity detection (BAD) deals with the task of predicting the presence or absence of bird vocalizations in a given audio recording. In this work, we propose an all-convolutional neural network (all-conv net) for bird activity detection. All the layers of this network including pooling and dense layers are implemented using convolution operations. The pooling operation implemented by convolution is termed as learned pooling. This learned pooling takes into account the inter featuremap correlations which are ignored in traditional max-pooling. This helps in learning a pooling function which aggregates the complementary information in various feature maps, leading to better bird activity detection. Experimental observations confirm this hypothesis. The performance of the proposed all-conv net is evaluated on the BAD Challenge 2017 dataset. The proposed all-conv net achieves state-of-art performance with a simple architecture and does not employ any data pre-processing or data augmentation techniques. This work is accepted for publication in INTERSPEECH 2018. feature_extract.py is used to extract melspectrogram features. all_convnet_BAD.py is the all convolutional model architecture all_convnet_BAD_maxpool_variant.py is the maxpool variant of the all_conv_net model get_activation_map.py is used to analyse any layer activations in the model To know more about BAD 2017 challenge and to download data, follow this link http://machine-listening.eecs.qmul.ac.uk/bird-audio-detection-challenge/

AnshThakur/CCSE_ICASSP

COMPRESSED CONVEX SPECTRAL EMBEDDING FOR BIRD SPECIES CLASSIFICATION

AnshThakur/continual_EHR_JACOB_A

Continual Learning of Electronic Health Records (EHR).

AnshThakur/DE

Directional embedding

AnshThakur/NC_Scores_Demo

AnshThakur/Quantum-Encoding

Quantum Encoding For Healthcare

AnshThakur/Self_aware_SGD

AnshThakur/All_conv_pytorch

# Pytorch: All Conv-Net for Bird Activity Detection Significance of Learned Pooling Bird activity detection (BAD)

AnshThakur/APE_MLSP2018

Archetypal Prototypal Embeddings (APE)

AnshThakur/DAA_IMK

AnshThakur/BAD_PSK

Rapid Bird Activity Detection Using Probabilistic Sequence Kernels

AnshThakur/Bird_seg_ICSPCS_2016

Model-based unsupervised segmentation of birdcalls from field recordings

AnshThakur/Bird_Species_Classification_Contrastive_loss

Bioacoustic classification using constrastive loss

AnshThakur/cifar-10-wrn

Using Wide Residual Networks to get state-of-the-art results in CIFAR-10 dataset

AnshThakur/CLA_MLSP2018

Convex likelihood alignments

AnshThakur/CNN_baselines

AnshThakur/Deep-Convex-Representation_IS2018

DCR_deep_archetypal_analysis

AnshThakur/DML

AnshThakur/MIMT-segmentation

AnshThakur/MLSP2017_bird_seg

AnshThakur/Self_Aware_SGD_INC

AnshThakur/shap-values

Shap values for model interpretation

AnshThakur/Test_Adv