Closed OPilgrim closed 1 year ago
Can you please check whether your environment setting is the same as we stated and whether the data is generated correctly?
I just re-evaluate again and get this:
Test: 100%|████████████████████████████████| 70/70 [01:13<00:00, 1.06s/it]
---------------------------------------------------------------------
Trigger I - P: 100.00 ( 403/ 403), R: 100.00 ( 403/ 403), F: 100.00
Trigger C - P: 100.00 ( 403/ 403), R: 100.00 ( 403/ 403), F: 100.00
---------------------------------------------------------------------
Role I - P: 75.93 ( 429/ 565), R: 76.47 ( 429/ 561), F: 76.20
Role C - P: 73.63 ( 416/ 565), R: 74.02 ( 416/ 562), F: 73.82
---------------------------------------------------------------------
It seems that both the gold trigger instance number (424) and the gold argument instance number (671) are different from mine.
Sorry, the above picture is ace05ep, this picture is ace05e.
And then this is my environment:
Package Version
----------------------- -----------
absl-py 1.2.0
asttokens 2.0.8
backcall 0.2.0
beautifulsoup4 4.9.3
bs4 0.0.1
cachetools 5.2.0
certifi 2022.9.24
charset-normalizer 2.1.1
click 8.1.3
decorator 5.1.1
executing 1.1.1
filelock 3.8.0
google-auth 2.12.0
google-auth-oauthlib 0.4.6
grpcio 1.49.1
idna 3.4
importlib-metadata 5.0.0
ipdb 0.13.9
ipython 8.5.0
jedi 0.18.1
joblib 1.2.0
lxml 4.6.3
Markdown 3.4.1
MarkupSafe 2.1.1
matplotlib-inline 0.1.6
numpy 1.23.3
oauthlib 3.2.1
packaging 21.3
parso 0.8.3
pexpect 4.8.0
pickleshare 0.7.5
Pillow 9.2.0
pip 22.2.2
prompt-toolkit 3.0.31
protobuf 3.19.6
ptyprocess 0.7.0
pure-eval 0.2.2
pyasn1 0.4.8
pyasn1-modules 0.2.8
Pygments 2.13.0
pyparsing 3.0.9
regex 2022.9.13
requests 2.28.1
requests-oauthlib 1.3.1
rsa 4.9
sacremoses 0.0.53
sentencepiece 0.1.95
setuptools 63.4.1
six 1.16.0
soupsieve 2.3.2.post1
stack-data 0.5.1
stanza 1.2
tensorboard 2.10.1
tensorboard-data-server 0.6.1
tensorboard-plugin-wit 1.8.1
tensorboardX 2.4
tokenizers 0.8.1rc2
toml 0.10.2
torch 1.8.0+cu111
torchaudio 0.8.0
torchvision 0.9.0+cu111
tqdm 4.64.1
traitlets 5.4.0
transformers 3.1.0
typing_extensions 4.4.0
urllib3 1.26.12
wcwidth 0.2.5
Werkzeug 2.2.2
wheel 0.37.1
zipp 3.9.0
Not sure what happens on your end. There is the results I got from re-inference:
Test: 100%|████████████████████████████████| 34/34 [00:38<00:00, 1.14s/it]
---------------------------------------------------------------------
Trigger I - P: 100.00 ( 424/ 424), R: 100.00 ( 424/ 424), F: 100.00
Trigger C - P: 100.00 ( 424/ 424), R: 100.00 ( 424/ 424), F: 100.00
---------------------------------------------------------------------
Role I - P: 75.04 ( 508/ 677), R: 75.93 ( 508/ 669), F: 75.48
Role C - P: 72.27 ( 490/ 678), R: 73.03 ( 490/ 671), F: 72.65
---------------------------------------------------------------------
I also update the readme a little bit and maybe you can reference the environment I put there to check whether there's any issue in your environment installation.
@OPilgrim Are you able to reproduce the results now? If yes, I'll close this issue.
OK, thanks very much! I'll try again later.
Dear Author, I used the
DEGREE_eae_ace05e.mdl
you provided and the scriptdegree/eval_pipelineEE.py -ceae config/ config_degree_eae_ace05e.json-eae [eae_model] -g
, ace05e was tested, and the following is the result. Why is it so low? Did I make a mistake?
which model do you use? base or large?
I finished this problem, but I forgot how to fixed it. lol. Thanks
@kk19990709 can you please double check your execution environment? Thanks
Dear Author, I used the![image](https://user-images.githubusercontent.com/24997865/195292615-4a8292f9-1634-4108-9e56-e9c46c2cb8d0.png)
DEGREE_eae_ace05e.mdl
you provided and the scriptdegree/eval_pipelineEE.py -ceae config/ config_degree_eae_ace05e.json-eae [eae_model] -g
, ace05e was tested, and the following is the result. Why is it so low? Did I make a mistake?