ThutmoseTaggerModel, a new model for inverse text normalization #4011
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
|
This pull request introduces 8 alerts when merging 42730ba into 9005f23 - view on LGTM.com new alerts:
|
added 4 commits
April 17, 2022 18:14
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
…icense headers Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
ekmb
requested changes
Apr 19, 2022
examples/nlp/text_normalization_as_tagging/conf/thutmose_tagger_itn_config.yaml
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/conf/thutmose_tagger_itn_config.yaml
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/conf/thutmose_tagger_itn_config.yaml
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/utils/eval_per_class.py
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/utils/prepare_corpora_after_alignment.py
Outdated
Show resolved
Hide resolved
ekmb
requested changes
Apr 19, 2022
examples/nlp/text_normalization_as_tagging/normalization_as_tagging_infer.py
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/normalization_as_tagging_infer.py
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/utils/extract_giza_alignments.py
Outdated
Show resolved
Hide resolved
| '--giza_suffix', type=str, required=True, help='suffix of alignment files, e.g. \"Ahmm.5\", \"A3.final\"' | ||
| ) | ||
| parser.add_argument('--out_filename', type=str, required=True, help='Output file') | ||
| parser.add_argument('--lang', type=str, required=True, help="Language") |
examples/nlp/text_normalization_as_tagging/utils/get_label_vocab.py
Outdated
Show resolved
Hide resolved
nemo/collections/nlp/models/text_normalization_as_tagging/thutmose_tagger.py
Show resolved
Hide resolved
nemo/collections/nlp/models/text_normalization_as_tagging/thutmose_tagger.py
Show resolved
Hide resolved
nemo/collections/nlp/models/text_normalization_as_tagging/thutmose_tagger.py
Outdated
Show resolved
Hide resolved
nemo/collections/nlp/models/text_normalization_as_tagging/thutmose_tagger.py
Outdated
Show resolved
Hide resolved
nemo/collections/nlp/data/text_normalization_as_tagging/thutmose_tagger_dataset.py
Show resolved
Hide resolved
added 3 commits
April 22, 2022 20:17
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
added 3 commits
April 28, 2022 00:35
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
|
This pull request introduces 1 alert when merging 1d48227 into 0d052c8 - view on LGTM.com new alerts:
|
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
|
This pull request introduces 1 alert when merging b03d10e into 70d9687 - view on LGTM.com new alerts:
|
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
ekmb
requested changes
Apr 28, 2022
examples/nlp/text_normalization_as_tagging/prepare_dataset_en.sh
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/prepare_dataset_en.sh
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/prepare_dataset_en.sh
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/utils/corpus_errors.ru
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/utils/extract_giza_alignments.py
Outdated
Show resolved
Hide resolved
examples/nlp/text_normalization_as_tagging/utils/filter_sentences_with_errors.py
Outdated
Show resolved
Hide resolved
ekmb
reviewed
Apr 28, 2022
nemo/collections/nlp/data/text_normalization_as_tagging/utils.py
Outdated
Show resolved
Hide resolved
|
|
||
| src_hiddens = self.bert_model(input_ids=input_ids, token_type_ids=segment_ids, attention_mask=input_mask) | ||
| log_softmax = self.logits(hidden_states=src_hiddens) | ||
| log_softmax_semiotic = self.semiotic_logits(hidden_states=src_hiddens) |
There was a problem hiding this comment.
why is this called log_softmax if log_softmax=False in the self.logits init?
| span_predictions.append(cid) | ||
| else: | ||
| span_predictions.append(self.tag_classification_report.num_classes - 1) # this stands for WRONG | ||
| assert len(span_labels) == len(span_predictions) |
There was a problem hiding this comment.
raise errors instead of assert
| else: | ||
| # this stands for WRONG | ||
| multiword_span_predictions.append(self.tag_classification_report.num_classes - 1) | ||
| assert len(multiword_span_labels) == len(multiword_span_predictions) |
There was a problem hiding this comment.
raise errors instead of assert
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
|
This pull request introduces 1 alert when merging fc8c6ce into 58ff608 - view on LGTM.com new alerts:
|
|
This pull request introduces 1 alert when merging 883d9e8 into 58ff608 - view on LGTM.com new alerts:
|
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
|
This pull request introduces 1 alert when merging 7073117 into 58ff608 - view on LGTM.com new alerts:
|
ekmb
reviewed
Apr 29, 2022
examples/nlp/text_normalization_as_tagging/dataset_preparation/prepare_corpora_for_alignment.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
ekmb
reviewed
Apr 29, 2022
examples/nlp/text_normalization_as_tagging/dataset_preparation/prepare_corpora_for_alignment.py
Show resolved
Hide resolved
nemo/collections/nlp/models/text_normalization_as_tagging/thutmose_tagger.py
Outdated
Show resolved
Hide resolved
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
ekmb
previously approved these changes
Apr 29, 2022
added 2 commits
April 30, 2022 00:40
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
ekmb
approved these changes
Apr 29, 2022
Signed-off-by: Alexandra Antonova <aleksandraa@nvidia.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Signed-off-by: Alexandra Antonova aleksandraa@nvidia.com
What does this PR do ?
A new tagger-based model for inverse text normalization
Collection: [NLP]
Changelog
Usage
# Add a code snippet demonstrating how to use thisBefore your PR is "Ready for review"
Pre checks:
PR Type:
If you haven't finished some of the above items you can still open "Draft" PR.
Who can review?
Anyone in the community is free to review the PR once the checks have passed.
Contributor guidelines contains specific people who can review PRs to various areas.
Additional Information