dataset load issue: I have replicated the code, however, after downloading the datasets (cub,emnist etc.), i am unable to train the model.

BayesWatch / deep-kernel-transfer

Official pytorch implementation of the paper "Bayesian Meta-Learning for the Few-Shot Setting via Deep Kernels" (NeurIPS 2020)

https://arxiv.org/abs/1910.05199

201 stars 29 forks source link

dataset load issue: I have replicated the code, however, after downloading the datasets (cub,emnist etc.), i am unable to train the model. #17

Closed Shantan243 closed 1 year ago

Shantan243 commented 2 years ago

I dont see .json files in the dataset, however, script requires the json files to train the model. iam i missing something.

base_file = configs.data_dir[params.dataset] + 'base.json' KeyError: 'C:\Users\shant_w5mrdz3\OneDrive\Desktop\deep-kernel-transfer-master\filelists\emnist\emnist+base.json'

mpatacchiola commented 2 years ago

Hi @Shantan243 you should download and preprocess the datasets by using the scripts in the repository:

cd filelists/DATASET_NAME/
sh download_DATASET_NAME.sh

Replace DATASET_NAME with one of the following: omniglot, CUB, miniImagenet, emnist, QMUL. Those scripts will download and create the json files you are missing.

See this section of the Readme for more details.

mpatacchiola commented 1 year ago

Closed for inactivity.