lichengunc

--[[
Implement beam search
]]
function layer:sample_beam(imgs, opt)
	local beam_size = utils.getopt(opt, 'beam_size', 10)
	local batch_size, feat_dim = imgs:size(1), imgs:size(2)
	local function compare(a,b) return a.p > b.p end -- used downstream
	local function compare_ppl(a, b) return a.ppl < b.ppl end  -- used upstream

	assert(beam_size <= self.vocab_size+1, 'lets assume this for now, otherwise this corner case causes a few headaches down the road. can be dealt with in future if needed')

lichengunc

import numpy as np
import torch
from torch.autograd import Variable
import torch.nn as nn
import torch.nn.functional as F

class RNNEncoder(nn.Module):
  def __init__(self, vocab_size, word_embedding_size, word_vec_size, hidden_size, bidirectional=False,
               input_dropout_p=0, dropout_p=0, n_layers=1, rnn_type='lstm', variable_lengths=True):
    super(RNNEncoder, self).__init__()

lichengunc

"""
LanguageRankingCriterion: takes [logp0, logp1] as input computing the ranking loss.
"""
class LanguageRankingCriterion(nn.Module):
  def __init__(self, margin=1.):
    super(LanguageRankingCriterion, self).__init__()
    self.margin = margin
  
  def forward(self, logprobs, target):
    """

lichengunc

  def _get_best_yaw_obj_from_pos(self, obj_id, grid_pos, height=1.0, use_iou=True):
    obj = self.objects[obj_id]
    obj_fine_class = obj['fine_class']

    cx, cy = self.env.house.to_coor(grid_pos[0], grid_pos[1])
    self.env.cam.pos.x = cx
    self.env.cam.pos.y = height
    self.env.cam.pos.z = cy

    best_yaw, best_coverage, best_mask = None, 0, None

lichengunc

def get_best_view_points(h3d, obj_id, args):
    # obj info
    obj = h3d.objects[obj_id]
    h3d.set_target_object(obj)
    obj_conn_map = h3d.env.house.connMapDict[obj_id][0]
    obj_point_cands = np.argwhere( (obj_conn_map > args.min_conn_dist) & (obj_conn_map <= args.max_conn_dist) )
    # don't search too many for saving time  
    if obj_point_cands.shape[0] > args.num_samples:
        perm = np.random.permutation(obj_point_cands.shape[0])[:args.num_samples]
        obj_point_cands = obj_point_cands[perm]

lichengunc

"""
This code will take torch.utils.data.Dataset as abstract class and implement the two important
functions:
1) __len__ : len(dataset) returns the size of the dataset
2) __getitem__: dataset[i] can be used to get i-th sample

data_json has 
0. refs:       [{ref_id, ann_id, box, image_id, split, category_id, sent_ids, att_wds}]
1. images:     [{image_id, ref_ids, file_name, width, height, h5_id}]
2. anns:       [{ann_id, category_id, image_id, box, h5_id}]

lichengunc

import cv2
import csv
import copy
import sys
import time
import pickle
import os
import os.path as osp
import itertools
import numpy as np

lichengunc

"""
We will load:
1) collected data = [{id, dataset, dataset_image_id, file_path, sent, split, bbox(xywh/None)}]
2) VQA questions = [{question, question_id, image_id}]
3) VG (on coco) questions = [{question, question_id, image_id}]
to make a full token_to_ix table.

Besides, we use the above to tokenize collected sentences, and make the follows:
1) token_to_ix   : token -> ix
2) split_to_iids : split -> img_ids

lichengunc

def tokenize(sent, token_to_ix=None):
  words = re.sub(r"([.,'!?\"()*#:;])", 
                 '', 
                 sent.lower()
                ).replace('-', ' ').replace('/', ' ').split()

  if token_to_ix:
    return [wd if wd in token_to_ix.keys() else 'UNK' for wd in words]
  else:
    return words