Distributed Methods
GPipe
Introduced by Huang et al. in GPipe: Efficient Training of Giant Neural Networks using Pipeline ParallelismGPipe is a distributed model parallel method for neural networks. With GPipe, each model can be specified as a sequence of layers, and consecutive groups of layers can be partitioned into cells. Each cell is then placed on a separate accelerator. Based on this partitioned setup, batch splitting is applied. A mini-batch of training examples is split into smaller micro-batches, then the execution of each set of micro-batches is pipelined over cells. Synchronous mini-batch gradient descent is applied for training, where gradients are accumulated across all micro-batches in a mini-batch and applied at the end of a mini-batch.
Source: GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism
Papers
| Paper | Code | Results | Date | Stars |
|---|
Tasks
| Task | Papers | Share |
|---|---|---|
| Image Classification | 2 | 16.67% |
| Machine Translation | 2 | 16.67% |
| Sentiment Analysis | 1 | 8.33% |
| Language Modelling | 1 | 8.33% |
| graph partitioning | 1 | 8.33% |
| BIG-bench Machine Learning | 1 | 8.33% |
| Link Prediction | 1 | 8.33% |
| Protein Folding | 1 | 8.33% |
| Fine-Grained Image Classification | 1 | 8.33% |
Usage Over Time
Components
| Component | Type |
|
|---|---|---|
| 🤖 No Components Found | You can add them if they exist; e.g. Mask R-CNN uses RoIAlign |
