search

omni-us / research-WriterAdaptation-HTR

Source code for WACV20 paper "Unsupervised Adaptation for Synthetic-to-Real Handwritten Word Recognition".
MIT License
15 stars 7 forks source link

load dataset #3

Open yanqi1811 opened 4 years ago

yanqi1811 commented 4 years ago

Hi, Could you elaborate on the data structure foe me? I want to train my own dataset, thank you! train_set = '/home/lkang/datasets/syn_60000_words/subset_20305words.txt' # 20305 words source domain valid_set = '/home/lkang/datasets/iam_final_words/RWTH.iam_word_gt_final.valid.thresh' test_set = '/home/lkang/datasets/iam_final_words/RWTH.iam_word_gt_final.test.thresh'

my source domain as train_set *****.jpg label target domain are only images, not labels.

Alikfp commented 2 years ago

Hi, Currently I am faced with the exact question that you had posted quite some time ago, I was wondering if you could figure it our eventually ?