Gantavya Bhatt

Inconsistent Selector gains and duplicate elements in Selector ranking in case of Facility Locations function in v0.6.0

6

Hi, I was working with two versions of Apricot, 0.5.0 and 0.6.0. I am using the Facility Locations Function, with a precomputed setting, by passing my own matrix. I have...

Discrepancy between the loss mentioned in the paper and GitHub

4

According to the paper, the negative component of the contrastive loss is the difference between the negative states (randomly sampled from embedding at timestamp t, (z_{t}~)) and the ground truth...

When instantiated model using diffusers, for ex. ```pipeline = StableDiffusionPipeline.from_pretrained('stabilityai/stable-diffusion-2-1-base', revision='main')``` Does it download the ema weights or non-ema weights?

Hi, Following tests fail when installing the software. ``` -- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html ====================================================================================================== short test summary info ======================================================================================================= FAILED tests/test_all.py::TestAll::test_dense_cpp_eval_groundset[GraphCut] - AssertionError: Eval on groundset is not >= 0 or...

Strange training dynamics for ImageReward model.

3

Hi! I am trying to train a reward model, and I am confused why in the initial iterations of training the gradients are not changing and neither the loss is...

Dataset download issue

2

Hi, I am trying to reproduce the 1k and 4k numbers for the ImageReward function accuracy, as mentioned in the paper. To do so, I downloaded the data, and modified...

Training ImageReward model on different budgets

5

Hi! The paper mentions that the training for the ImageReward model is not easy and is sensitive to hyperparameters. In the section about hyperparameters, it says --" We find that...