Eric Hofesmann
Eric Hofesmann
Currently for datasets with bounding boxes, we need to specify the max bounding boxes possible so all output batches are of the same size: https://github.com/MichiganCOG/ViP/blob/74776f2575bd5339ba39c784bbda4f04cc859add/datasets/ImageNetVID.py#L27 What we should do is...
Allow you to specify the exact pytorch class and related parameters in the yaml file for the optimizer and scheduler you want to run. (e.g. `torch.schedulers.MultiStepLR`)
Create a class object to be passed through every model, loss, and metric that has a method allowing you to add a plot to tensorboard for any specified variable.
The example from the [TF2 Model Zoo inference tutorial notebook](https://github.com/tensorflow/models/blob/master/research/object_detection/colab_tutorials/inference_tf2_colab.ipynb) loads images in RGB format with: ``` def load_image_into_numpy_array(path): """Load an image from file into a numpy array. Puts image...
Users have reported that [`merge_samples()`](https://voxel51.com/docs/fiftyone/api/fiftyone.core.dataset.html?highlight=merge%20samples#fiftyone.core.dataset.Dataset.merge_samples) can occasionally fail to merge all samples, most noticeably for large video datasets. Work needs to be done to replicate this, but the setup is...
When drawing labels on samples and saving them to disk, it is expected that unique filenames will be created if multiple samples in the view have the same filename/path. However,...
Dataset names should be validated for unsupported characters before being saved. Many non-alphanumeric characters result in issues when attempting to load the dataset in the App. ```python import fiftyone as...
Following https://github.com/voxel51/fiftyone/pull/1858, the Lightning Flash integration needs to be updated to allow the reference to `default_classes` to be removed from the [Lightning Flash docs](https://github.com/voxel51/fiftyone/blob/develop/docs/source/integrations/lightning_flash.rst) discussed in this thread https://github.com/voxel51/fiftyone/pull/1953. Note...
When making calls to the CVAT REST API, some of the requests can raise errors due to server issues requiring methods like dataset.annotate() and dataset.load_annotations() to be run multiple times...