Transformer ELMo config fixes (#131)

* allow multiple token embedders, but only use first with non-empty type

* blacken the modified code

* update CHANGELOG

* Throw an exception if we already know the model isn't going to work.

* Update the transformer language model config.

* Changelog

* Brings back the test for TransformerElmo

* Don't reformat a file that's full of tables

* Formatting

Co-authored-by: Anthony P Panza <[email protected]>

CHANGELOG.md

@@ -15,6 +15,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Added more information to model cards for pair classification models (`pair-classification-decomposable-attention-elmo`, `pair-classification-roberta-snli`, `pair-classification-roberta-mnli`, `pair-classification-esim`).
 
+### Fixed
+
+- Fixed TransformerElmo config to work with the new AllenNLP
+
+
 ## [v1.2.0](https://github.com/allenai/allennlp-models/releases/tag/v1.2.0) - 2020-10-29
 
 ### Changed
@@ -30,6 +35,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Fixed minor bug in MaskedLanguageModel, where getting token ids used hard-coded assumptions (that
   could be wrong) instead of our standard utility function.
 
+
 ## [v1.2.0rc1](https://github.com/allenai/allennlp-models/releases/tag/v1.2.0rc1) - 2020-10-22
 
 ### Added
@@ -62,6 +68,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
   the desired behavior is to not add start or end symbols to either the source or the target
   and the default `start_symbol` or `end_symbol` are not part of the tokenizer's vocabulary.
 
+
 ## [v1.1.0](https://github.com/allenai/allennlp-models/releases/tag/v1.1.0) - 2020-09-08
 
 ### Fixed
@@ -70,6 +77,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Fixed evaluation of metrics when using distributed setting.
 - Fixed a bug introduced in 1.0 where the SRL model did not reproduce the original result.
 
+
 ## [v1.1.0rc4](https://github.com/allenai/allennlp-models/releases/tag/v1.1.0rc4) - 2020-08-21
 
 ### Added
@@ -78,6 +86,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Added a test for the pretrained sentiment analysis model.
 - Added way for questions from quora dataset to be concatenated like the sequences in the SNLI dataset.
 
+
 ## [v1.1.0rc3](https://github.com/allenai/allennlp-models/releases/tag/v1.1.0rc3) - 2020-08-12
 
 ### Fixed
@@ -87,7 +96,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Made the SST reader a little more strict in the kinds of input it accepts.
 
 
-
 ## [v1.1.0rc2](https://github.com/allenai/allennlp-models/releases/tag/v1.1.0rc2) - 2020-07-31
 
 ### Changed
@@ -110,7 +118,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [v1.1.0rc1](https://github.com/allenai/allennlp-models/releases/tag/v1.1.0rc1) - 2020-07-14
 
-
 ### Fixed
 
 - Updated the BERT SRL model to be compatible with the new huggingface tokenizers.
@@ -132,10 +139,12 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Added a category for multiple choice models, including a few reference implementations
 - Implemented manual distributed sharding for SNLI dataset reader.
 
+
 ## [v1.0.0](https://github.com/allenai/allennlp-models/releases/tag/v1.0.0) - 2020-06-16
 
 No additional note-worthy changes since rc6.
 
+
 ## [v1.0.0rc6](https://github.com/allenai/allennlp-models/releases/tag/v1.0.0rc6) - 2020-06-11
 
 ### Changed
@@ -157,6 +166,7 @@ No additional note-worthy changes since rc6.
 - Added sentiment analysis models to pretrained.py
 - Added NLI models to pretrained.py
 
+
 ## [v1.0.0rc5](https://github.com/allenai/allennlp-models/releases/tag/v1.0.0rc5) - 2020-05-14
 
 ### Changed
@@ -171,6 +181,7 @@ No additional note-worthy changes since rc6.
 
 - `conllu` dependency (previously part of `allennlp`'s dependencies)
 
+
 ## [v1.0.0rc4](https://github.com/allenai/allennlp-models/releases/tag/v1.0.0rc4) - 2020-05-14
 
 We first introduced this `CHANGELOG` after release `v1.0.0rc4`, so please refer to the GitHub release

test_fixtures/lm/bidirectional_language_model/training_data/sentences1.txt

@@ -0,0 +1,2 @@
+I run the unit test thrice.
+I run the unit test four times.

test_fixtures/lm/bidirectional_language_model/training_data/sentences2.txt

@@ -0,0 +1,2 @@
+I run the unit test once.
+I run the unit test twice.

test_fixtures/lm/bidirectional_language_model/vocab/non_padded_namespaces.txt

@@ -0,0 +1 @@
+*labels\n*tags