20,290
edits
Assembly Line Activity Recognition: Difference between revisions
m
→Network implementation
mNo edit summary |
|||
| Line 202: | Line 202: | ||
The network used is a PyTorch implementation from the PyTorchVideo library. The following figure shows a basic representation of the layers that constitute this network. | The network used is a PyTorch implementation from the PyTorchVideo library. The following figure shows a basic representation of the layers that constitute this network. | ||
<br> | |||
<br> | |||
[[File:Slowfast_assembly_network.png|500px|center|thumb|SlowFast network basic representation]] | [[File:Slowfast_assembly_network.png|500px|center|thumb|SlowFast network basic representation]] | ||
<br> | |||
This project was built around this PyTorchVideo SlowFast implementation; the software allows us to experiment with this network for our specific use case, modify parameters and test multiple training and tuning techniques in order to achieve the best results possible. For experiment tracking and reproducibility this project was built using DVC as MLOps tool, with a 5 stage process as seen in the following dependency diagram. | This project was built around this PyTorchVideo SlowFast implementation; the software allows us to experiment with this network for our specific use case, modify parameters and test multiple training and tuning techniques in order to achieve the best results possible. For experiment tracking and reproducibility this project was built using DVC as MLOps tool, with a 5 stage process as seen in the following dependency diagram. | ||