lx

kitti_seg_input.py

1

'arch' in the following should be 'jitter', right? ``` def crop_to_size(hypes, image, gt_image): new_width = image.shape[1] new_height = image.shape[0] width = hypes['arch']['image_width'] height = hypes['arch']['image_height'] ```

[Question] why does RerankingEvaluator implementation use embeddings + cos_sim instead of using similarity score from model?

1

as far as i understand, reranker models, typcially take query + doc as input, and directly give a score as output. so called "cross-encoders".  however, when i read RerankingEvaluator...

这个只能在windows 下运行？看到 data目录有些 exe 文件。

1

linux + conda 可以不？