mgh1
mgh1
Hi Sergey - I noticed in /vislab/feature.py there is a compute command which appears to load the features. I'd just like to confirm the end-to-end procedure of adding my own...
It seems the main method to load Caffe's CNN features is from the caffe function in vislab / features / misc.py. I have several questions on performance: 1. Why does...
## ❓ Questions and Help ### Our Use-Case Thanks to the authors/contributors for such an amazing project! Our use-case is to use Captum to help visualize the word importance for...
**Describe the bug** On the official documentation website [here](https://simpletransformers.ai/docs/classification-specifics/), it lists Longformer with a * under the `Supported Model Types` section. Then a legend with * says `* Not available...
**Is your feature request related to a problem? Please describe.** Longformer is a popular model type but not supported in [`RepresentationModel`](https://github.com/ThilinaRajapakse/simpletransformers/blob/a44487d36ce56e329b0e03ac9dc369c42233bcee/simpletransformers/language_representation/representation_model.py) , only `roberta`, `bert`, and `gp2` are supported. Please...
I noticed using the same dataset for test and eval? Why do this and not separate out an eval set? ```python def test_dataloader(self): """test dataloader""" return DataLoader( self.test_dataset, batch_size=self.batch_size, shuffle=False,...
In the `predict` function there is a `do_sample` argument. It apparently does nothing because it is not used at all in the function.
Has anyone done a comparison test that shows the Niuzhiheng-Caffe and BLVC-Caffe versions produce the same accuracy on a comprehensive task, say the ImageNet classification task? It would be great...