Open Issues Need Help
View All on GitHub pxt_torch_dataset_collate_fn should have a more generic name and use pytorch default collate for all keys in conversion_dict about 23 hours ago
good first issue
Use hot encoding for label informations and keep the labels value in an informative columns about 24 hours ago
help wanted
good first issue
good first issue in progress
GoldFeatureExtractor: make the device always available (allow to convert data device before running extraction) 8 months ago
good first issue
good first issue
GoldSplitter: must force the selector to find features to vectorize in the "feature" column 8 months ago
good first issue
GoldDescriptor, GoldSelector and GoldSplitter: Allow to specify to restrict to a small part of the dataset 8 months ago
good first issue
GoldDescriptor: Add a data_key attribute to find which element of the dataset to extract feature from 8 months ago
enhancement good first issue
GoldSplitter and GoldSelector: When multiple classes are selected we should allow to name the table differently 8 months ago
good first issue help wanted
good first issue
good first issue
help wanted
GoldDescriptor, GoldSelector, GoldSplitter: show progress and make pixeltable logging optional 8 months ago
help wanted