new-pos-rel-stats.py

confusion_dict_pos = {}
confusion_dict_paths = {}

# NEW STUFF #
addition_stats_pos = Counter()
addition_stats_rel = Counter()
# NEW STUFF #

strip_direction = lambda x: x.split('_')[0]

for i in range(len(en)):
    en_n, en_g = conll2graph(en[i])
    ko_n, ko_g = conll2graph(ko[i])
    alignment = alignments[i]
    # Simplify the alignment to a set of one-to-one pairs
    one_to_one = []
    for k, v in alignment.items():
        if k == 'X':
            # NEW STUFF #
            for idx in v:
                if '.' in idx:
                    continue
                new_node = ko_n[idx]
                addition_stats_pos[new_node['pos']] += 1
                addition_stats_rel[new_node['relation']] += 1
            # NEW STUFF #
        head = k
        tail = str(highest_or_none(v, ko_g))
        one_to_one.append((head, tail))
    # POS confusion dict
    for pair in one_to_one:
        head, tail = pair
        # Skip technical additional nodes
        if '.' in head:
            continue
        try:
            en_pos = en_n[head]['pos']
        except KeyError:
            print(i, en[i])
            continue
        if tail == 'None':
            ko_pos = 'None'
        else:
            ko_pos = ko_n[tail]['pos']
        if en_pos not in confusion_dict_pos:
            confusion_dict_pos[en_pos] = Counter()
        confusion_dict_pos[en_pos][ko_pos] += 1
    # Path confusion dict
    for pair in combs(one_to_one, 2):
        (en_head, ko_head), (en_tail, ko_tail) = pair
        # Skip technical additional nodes
        if '.' in head:
            continue
        en_path_arr = get_path(en_head, en_tail, en_g)
        if len(en_path_arr) > 1:
            continue
        en_path = strip_direction(en_path_arr[0])
        if ko_head == ko_tail:
            ko_path = 'Nodes collapsed'
        elif ko_head == 'None' and ko_tail == 'None':
            ko_path = 'Both endpoints unaligned'
        elif ko_head == 'None' or ko_tail == 'None':
            ko_path = 'One endpoint unaligned'
        else:
            ko_path_arr = get_path(ko_head, ko_tail, ko_g)
            ko_path = '->'.join(
                list(map(strip_direction, ko_path_arr))
            )
        if en_path not in confusion_dict_paths:
            confusion_dict_paths[en_path] = Counter()
        confusion_dict_paths[en_path][ko_path] += 1