xmodaler.functional

xmodaler.functional.boxes_to_locfeats(boxes, image_w, image_h)

xmodaler.functional.caption_to_mask_tokens(caption, max_seq_length, tokenizer, need_g_tokens=False, need_no_mask_tokens=False, must_mask=False)

xmodaler.functional.decode_sequence(vocab, seq)

xmodaler.functional.decode_sequence_bert(tokenizer, seq, sep_token_id)

xmodaler.functional.dict_as_tensor(input_dict)

xmodaler.functional.dict_to_cuda(input_dict)

xmodaler.functional.expand_tensor(tensor, size, dim=1)

xmodaler.functional.clip_v_inputs(v_feats, spatials, image_mask)

xmodaler.functional.clip_t_inputs(input_txt, segment_ids, input_mask)

xmodaler.functional.iou(anchors, gt_boxes)

anchors: (N, 4) ndarray of float gt_boxes: (K, 4) ndarray of float overlaps: (N, K) ndarray of overlap between boxes and query_boxes

xmodaler.functional.load_vocab(path)

xmodaler.functional.pad_tensor(tensor, padding_value, use_mask)

xmodaler.functional.random_region(image_feats, overlaps)

xmodaler.functional.random_word(tokens, tokenizer, must_mask=False)

xmodaler.functional.read_lines(path)

xmodaler.functional.read_lines_set(path)

xmodaler.functional.read_np(path)

xmodaler.functional.read_np_bbox(path, max_feat_num, use_global_v=True)

xmodaler.functional.flat_list_of_lists(l)

flatten a list of lists [[1,2], [3,4]] to [1,2,3,4]