luheng / lsgn Goto Github PK
View Code? Open in Web Editor NEWLabeled Span Graph Networks
License: Apache License 2.0
Labeled Span Graph Networks
License: Apache License 2.0
I am using the version for python3 (with tenserflow==1.14.0). I am trying to test the pre-trained model on the sample file, and being on the last stage with decoder.py I got this error:
Traceback (most recent call last):
File "decoder.py", line 38, in
model = SRLModel(data, config)
File "/Users/nikahelicopter/Desktop/GERMANY/DR.SOLOPOVA/lsgn-python3/srl_model.py", line 24, in init
self.data.input_dict, self.data.labels_dict)
File "/Users/nikahelicopter/Desktop/GERMANY/DR.SOLOPOVA/lsgn-python3/srl_model.py", line 108, in get_predictions_and_loss
max_sentence_length, sort_spans=False, enforce_non_crossing=False)
File "/Users/nikahelicopter/Desktop/GERMANY/DR.SOLOPOVA/lsgn-python3/model_utils.py", line 212, in get_batch_topk
sort_spans, enforce_non_crossing) # [num_sentences, max_num_predictions]
File "", line 89, in extract_spans
File "/Users/nikahelicopter/Library/Python/3.6/lib/python/site-packages/tensorflow/python/framework/op_def_library.py", line 788, in _apply_op_helper
op_def=op_def)
File "/Users/nikahelicopter/Library/Python/3.6/lib/python/site-packages/tensorflow/python/util/deprecation.py", line 507, in new_func
return func(*args, **kwargs)
File "/Users/nikahelicopter/Library/Python/3.6/lib/python/site-packages/tensorflow/python/framework/ops.py", line 3616, in create_op
op_def=op_def)
File "/Users/nikahelicopter/Library/Python/3.6/lib/python/site-packages/tensorflow/python/framework/ops.py", line 1952, in init
raise ValueError("'%s' is not a valid node name" % node_def.name)
ValueError: '' is not a valid node name
Can it have something to do with this comment? # TODO: Make labels_dict = None at test time.
I have to change something as I use it in test mode?
I also wonder why in build_custom_kernels.sh for mac we use coref_kernels.so and with windows we build srl_kernels.so?
I really am looking forward to your answer, as my troubleshooting does not help much for now:(
Thank you in advance!
Veronika
Traceback (most recent call last):
File "decoder.py", line 12, in
from lsgn_data import LSGNData
File "/hltc/0/cl/tools/srl_parsers/l_he_etal_2018/lsgn/lsgn_data.py", line 2, in
import tensorflow_hub as hub
File "/homes/skumyol/venv2.7/lib/python2.7/site-packages/tensorflow_hub/init.py", line 29, in
from tensorflow_hub.estimator import LatestModuleExporter
File "/homes/skumyol/venv2.7/lib/python2.7/site-packages/tensorflow_hub/estimator.py", line 25, in
from tensorflow_hub import tf_utils
File "/homes/skumyol/venv2.7/lib/python2.7/site-packages/tensorflow_hub/tf_utils.py", line 33, in
from tensorflow.python.feature_column import feature_column_v2
ImportError: cannot import name feature_column_v2
File "decoder.py", line 14, in
from srl_model import SRLModel
File "/Users/Desktop/lsgn-master/srl_model.py", line 15, in
from model_utils import *
File "/Users/Desktop/lsgn-master/model_utils.py", line 5, in
import srl_ops
File "/Users/Desktop/lsgn-master/srl_ops.py", line 4, in
srl_op_library = tf.load_op_library("./srl_kernels.so")
File "/Users/opt/anaconda3/lib/python3.7/site-packages/tensorflow_core/python/framework/load_library.py", line 61, in load_op_library
lib_handle = py_tf.TF_LoadLibrary(library_filename)
tensorflow.python.framework.errors_impl.NotFoundError: dlopen(./srl_kernels.so, 6): image not found
Stuck here. Would be great if someone could help!
I cannot find it using a simple google search
Hi, the address of srlconll dataset is not accessible for me, http://www.lsi.upc.edu/~srlconll/srlconll-1.1.tgz
can anybody tell me how to download the dataset? thank you.
It seem that cat can't handle the large number of *.gold_conll in the train portion. The script print this warning and continue:
./scripts/make_conll2012_data.sh: line 19: /bin/ls: Argument list too long
However, changing:
lsgn/scripts/make_conll2012_data.sh
Line 19 in b1495ec
lsgn/scripts/make_conll2012_data.sh
Line 20 in b1495ec
to:
for file in ${ONTONOTES_PATH}/data/train/data/english/annotations/*/*/*/*.gold_conll
do
cat "$file" >> ${SRL_PATH}/train.english.v5_gold_conll
done
solve the problem
Pyhon 2.7 with tf 1.8 is giving such error.
Thanks for your admiring work!
But the following files are not provided, could you provide these files?
elmo/english.lm_embeddings.skip.hdf5
elmo/ontonotes5.test.english.lm_embeddings.skip.hdf5
elmo/ontonotes5.train.english.lm_embeddings.skip.hdf5
elmo/ontonotes5.dev.english.lm_embeddings.skip.hdf5
elmo/conll05.train.elmo_embeddings.hdf5
elmo/conll05.dev.elmo_embeddings.hdf5
elmo/conll05.test_wsj.elmo_embeddings.hdf5
elmo/conll05.test_brown.elmo_embeddings.hdf5
I tried hundreds times of training the model with singleton.py, the processing always suspend at:
tf_loss, tf_global_step, _ = session.run([model.loss, model.global_step, model.train_op])
you see from the snapshot that the codes is ok to run but, is suspended at there.
After several debug, I will say the problem is located on the:
enqueue_thread = threading.Thread(target=_enqueue_loop)
since the process never get into the '_enqueue_loop()'.
Other word, the 'FIFOQueue' with 'thread' failed to schedule.
Help with this plz.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.