An Efficient Recurrent Unit (ERU) extends LSTM-based language models by replacing linear transforms for processing the input vector with the EESP unit inside the LSTM cell.
Source: ESPNetv2: A Light-weight, Power Efficient, and General Purpose Convolutional Neural NetworkPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
document understanding | 1 | 11.11% |
Visual Grounding | 1 | 11.11% |
General Classification | 1 | 11.11% |
Image Classification | 1 | 11.11% |
Language Modelling | 1 | 11.11% |
Object Detection | 1 | 11.11% |
Real-Time Object Detection | 1 | 11.11% |
Real-Time Semantic Segmentation | 1 | 11.11% |
Semantic Segmentation | 1 | 11.11% |
Component | Type |
|
---|---|---|
Dropout
|
Regularization | |
EESP
|
Image Model Blocks | |
LSTM
|
Recurrent Neural Networks |