Skip to content

"Evaluate current model using CoNLL script" does no run #2

@XinyiYan-njust

Description

@XinyiYan-njust

It seems like there is somthing wrong with → “os.system("%s < %s > %s"%(eval_script, output_path, scores_path))”
the files in output_path have text however the files in scores_path are null, which lead to the list of eval_lines being null.

The error is showing below:
'evaluation' �����ڲ����ⲿ���Ҳ���ǿ����еij���
�����������
Traceback (most recent call last):
File "C:/Users/70962/PycharmProjects/named-entity-recognition/ner-tagger/train.py", line 255, in
dev_data, id_to_tag, dico_tags, epoch)
File "C:\Users\70962\PycharmProjects\named-entity-recognition\ner-tagger\utils.py", line 353, in evaluate
print(eval_lines[1])
IndexError: list index out of range

BTW input parameters are
--train
../data/train/train.txt
--dev
../data/dev/dev.txt
--test
../data/test/test.txt
--eval
evaluation/temp
--model
model
--tag_scheme
iob
--zeros
1
--with_eeg_gaze
1
--bins
100
--fold
50

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions