Akshay Dudhane

icon company Mohomman bin Zayed University of Artificial Intelligence icon location Masdar City, Abu Dhabi icon location 🌟 Research Scientist | Machine Learning Enthusiast | Open Source Contributor

Results 3 comments of Akshay Dudhane

About batch size in training

Yes, you understood correctly. As burst is of 4 dimensions, we don't have any alternative other than keeping batch size 1. One thing you can try is to combine burst...

Question for training time

For synthetic burst SR, we train the proposed Burstormer for 300 epochs with 4 RTX6000 GPUs, and it takes roughly around 5-6 days. To fine-tune this model for real burst...

Question for training time

Burstormer is faster than BIPNet. In our settings, we keep precision 16 and deterministic True. In released codes, each GPU serves a single burst (so 4 GPUs combinely make batch...