A new codebase for popular Scene Graph Generation methods (2020). Visualization & Scene Graph Extraction on custom images/datasets are provided. It's also a PyTorch implementation of paper “Unbiased Scene Graph Generation from Biased Training CVPR 2020”
MIT License
1.06k
stars
229
forks
source link
How to Train with Epochs in Addition to Iterations? #172
I noticed that in the codebase (https://github.com/KaihuaTang/Scene-Graph-Benchmark.pytorch/blob/master/tools/relation_train_net.py#L134), the training logic goes by a single pass through the train loader with a max of 16000 batches. How to modify the code to run epochs on top of the iterations or just manually ran train.sh again and again? I have tried to do this myself but I'm not sure if this is already implemented and I've missed it or there are additional concerns I have overlooked.
❓ Questions and Help
Hi Kaihua and the SGG community,
I noticed that in the codebase (https://github.com/KaihuaTang/Scene-Graph-Benchmark.pytorch/blob/master/tools/relation_train_net.py#L134), the training logic goes by a single pass through the train loader with a max of 16000 batches. How to modify the code to run epochs on top of the iterations or just manually ran train.sh again and again? I have tried to do this myself but I'm not sure if this is already implemented and I've missed it or there are additional concerns I have overlooked.