frank-xwang / RIDE-LongTailRecognition

[ICLR 2021 Spotlight] Code release for "Long-tailed Recognition by Routing Diverse Distribution-Aware Experts."

frank-xwang/RIDE-LongTailRecognition Issues

about multi-experts network
Closed a year ago
The training accuracy of imagenet data set is not up to standard
Closed a year ago6
[GPU Utilization] DataLoader iteration speed quite low at the start of every epoch
Closed 2 years ago2
How to understand Formula 13 and 14 in your paper?
Closed 2 years ago2
[Error] An error that may cause the "reweight" flag to fail
Closed 2 years ago1
Question about Diversity loss
Closed 2 years ago2
[Error] An pRuntimeError: invalid argument 5: k not in range for dimension at C:/w/b/windows/pytorch/aten/src\THC/generic/THCTensorTopK.cu:26ython exception is encountered at ...
Closed 2 years ago2
the imagenet-lt dataset is note available
Closed 2 years ago1
LDAM loss implementation issue
Closed 2 years ago1
Request for more results
Closed 3 years ago1
[Error] n_gpu model encounted errors
Closed 3 years ago2
Mismatched hyper-parameter settings for
Closed 3 years ago2
Question on the function of the normalized linear layer
Closed 3 years ago1
[Error] RIDE Expert Assignment Module Training (Stage 2)
Closed 3 years ago2
How to implement cRT and τ-norm with this framework?
Closed 3 years ago2
About the change of network structure
Closed 3 years ago7
About routing loss
Closed 3 years ago1
Can you release the resulting model from the first phase of training?
Closed 3 years ago2
CIFAR 10 support?
Closed 3 years ago1
The diversity loss has no effect?
Closed 3 years ago5
Could you release the code that measure the bias and variance?
Closed 3 years ago1
[Error] An python exception is encountered at stage 2
Closed 3 years ago5
[Conceptual question] Are RIDE and EA used at the same time?
Closed 3 years ago1
When will the pertained models be released?
Closed 3 years ago2
where is the routing module in the code?
Closed 3 years ago2
the position of experts
Closed 3 years ago1
[Conceptual question] Training Stage 1 only is better than full training when # experts is 4
Closed 3 years ago1
about temperature_mean
Closed 3 years ago1
[Conceptual question] Question about self-distillation
Closed 3 years ago1
The mismatch between your LDAM implementation and the original one
Closed 3 years ago1
The mismatch between code and paper
Closed 3 years ago1
Hello, when can the code be released?
Closed 3 years ago2