Robik Shrestha
Robik Shrestha
Hi, Would it be possible to release the text-to-image retrieval results file? Thanks!
Could you provide ids for the train/val splits for SBU? I do not need the full db, just want to ensure I am using the same splits. Also, was wondering...
Hi, The colored MNIST accuracies differ a lot for different random seeds, set using `torch.manual_seed(seed)`. **Without random seed:** before resampling (49%), after resampling (57%) **With random seed set to 1:**...
Hi, I had a question regarding the PyTorch implementation of LearnedMixin. https://github.com/chrisc36/debias/blob/af7f0e40f9120ae2d3081cb8a2bf4dad64a18aa7/debias/bert/clf_debias_loss_functions.py#L41 ``` def forward(self, hidden, logits, bias, labels): logits = logits.float() # In case we were in fp16 mode...
How many output units do you use for age classification? Is each age a separate class? Or is it a binary problem 0-29 vs 40+?
Hi, In the [_train_step function](https://github.com/feidfoe/learning-not-to-learn/blob/fad2ec8faab52c2191d82e15164ab7b968009682/trainer.py#L106), the code performs two forward passes: 1) to compute main loss and 'pseudo loss' for biases 2) to compute classification loss for biases. Isn't a...
I am using the project you created and testing it using JMeter. I have a view scoped bean accessed by xhtml. From JMeter, I am sending continuous requests. I find...
Hi, After trying to run the code for CelebA (with ResNet18), I get the following results: 1. valid_attrwise_accs_b ` tensor([[0.9578, 0.9966], [0.8382, 0.2747]]) ` 2. valid_attrwise_accs_d `tensor([[0.0300, 0.0205], [0.6841, 0.9231]])...
Would it be possible to provide `trainval_ans2label.pkl` and `trainval_label2ans.pkl` files. The files generated from compute_softscore.py does not seem to be consistent with the labels in `VQA_caption_{name}dataset.pkl` files. For instance, for...