microsoft / icecaps Goto Github PK

Are there any plans to add support for tf2? Migrating (into a new parallel branch maybe?) should not be super hard. I could work on a PR for some of the migrations if there are any plans to shift soon?

create_dict.py text2tfrecord.py files seems to be missing.

While I was trying to generate dataset by executing GenMultiSpeakerDataset, it tried to execute create_dict.py text2tfrecord.py files. Which I couldn't find in the repo.

Hi all, thanks for the repo!

In the paper, you have mentioned you would provide pretrained personality embeddings that can be used for applications. May I ask if this indeed shared somewhere in the repo or if it can be shared with us via dropbox or gdrive? Thanks.

There are important files that Microsoft projects should all have that are not present in this repository. A pull request has been opened to add the missing file(s). When the pr is merged this issue will be closed automatically.

Microsoft teams can learn more about this effort and share feedback within the open source guidance available internally.

Merge this pull request

I'm curious on this project and I am installing it within a container with all needed components. I have run into an error after entering a query at the query prompt. For awareness, I'm new to python and deep learning.

Query: Why do you error?
Traceback (most recent call last):
File "train_simple_example.py", line 71, in
tf.app.run()
File "/opt/conda/lib/python3.7/site-packages/tensorflow/python/platform/app.py", line 125, in run
_sys.exit(main(argv))
File "train_simple_example.py", line 65, in main
decoding.cmd_decode(model, vocab, persona=True)
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 319, in cmd_decode
cmd_listener, model, vocab, sys.stdout, mmi_component, persona, max_len, lambda_balance, num_turns, eos_token)
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 313, in interactive_decode
return decode(model, input_fn_src, vocab, outstream=(None if mmi_component else outstream), hooks=[hook], save_all=(mmi_component is not None))
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 104, in decode
for pred in predictions:
File "/opt/conda/lib/python3.7/site-packages/icecaps/estimators/abstract_icecaps_estimator.py", line 296, in predict
checkpoint_path=(checkpoint_path or self.best_ckpt), yield_single_examples=yield_single_examples)
File "/opt/conda/lib/python3.7/site-packages/tensorflow_estimator/python/estimator/estimator.py", line 627, in predict
hooks=all_hooks) as mon_sess:
File "/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/monitored_session.py", line 934, in init
stop_grace_period_secs=stop_grace_period_secs)
File "/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/monitored_session.py", line 648, in init
self._sess = _RecoverableSession(self._coordinated_creator)
File "/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/monitored_session.py", line 1122, in init
_WrappedSession.init(self, self._create_session())
File "/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/monitored_session.py", line 1127, in _create_session
return self._sess_creator.create_session()
File "/opt/conda/lib/python3.7/site-packages/tensorflow/python/training/monitored_session.py", line 812, in create_session
hook.after_create_session(self.tf_sess, self.coord)
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 249, in after_create_session
self._get_feed_dict())
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 226, in _get_feed_dict
features = self._input_fn(self, context)
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 309, in _fn_get_input
return convert_interactive_input(in_msg, context, vocab, pad_len=max_len)
File "/opt/conda/lib/python3.7/site-packages/icecaps/decoding/decoding.py", line 293, in convert_interactive_input
in_msg = AbstractDataProcessor.basic_preprocess(in_msg)
File "/opt/conda/lib/python3.7/site-packages/icecaps/data_io/abstract_data_processor.py", line 48, in basic_preprocess
line = DataProcessor.tokenize_punctuation(line)
NameError: name 'DataProcessor' is not defined

In the paper "Novel positional encodings to enable tree-based transformers", this repo was referred to. But I didn't find the example code for this part, seems this repo for the chatbot work.

Could I know where is the repo for that paper?