Fonduer

-------------------

Fixed
^^^^^
* `lukehsiao`_: Updated snorkel-metal version requirement to ensure new syntax
works when a user upgrades Fonduer.
* `lukehsiao`_: Improve error messages on PostgreSQL connection and update FAQ.

-------------------

Added
^^^^^
* `senwu`_: Add ``SparseLSTM`` support reducing the memory used by the LSTM
for large applications.
(`175 <https://github.com/HazyResearch/fonduer/pull/175>`_)

.. note::
With the SparseLSTM discriminative model, we save memory for the origin
LSTM model while sacrificing runtime. In Fonduer v0.3.5, SparseLSTM is as
follows:

.. code:: python

from fonduer.learning import SparseLSTM

disc_model = SparseLSTM()
disc_model.train(
(train_cands, train_feature), train_marginals, n_epochs=5, lr=0.001
)

Fixed
^^^^^
* `senwu`_: Fix issue with ``get_last_documents`` returning the incorrect
number of docs and update the tests.
(`176 <https://github.com/HazyResearch/fonduer/pull/176>`_)

* `senwu`_: Use the latest MeTaL syntax and fix flake8 issues.
(`173 <https://github.com/HazyResearch/fonduer/pull/173>`_)

-------------------

Changed
^^^^^^^
* `senwu`_: Use ``sqlalchemy`` to check connection string. Use ``postgresql``
instead of ``postgres`` in connection string.

Fixed
^^^^^
* `lukehsiao`_: The features/labels/gold_label key tables were not properly
designed for multiple relations in that they indistinguishably shared the
global index of keys. This fixes this issue by including the names of the
relations associated with each key. In addition, this ensures that clearing a
single relation, or relabeling a single training relation does not
inadvertently corrupt the global index of keys.
(`167 <https://github.com/HazyResearch/fonduer/pull/167>`_)

-------------------

Changed
^^^^^^^
* `lukehsiao`_: Added ``longest_match_only`` parameter to
:class:`LambdaFunctionMatcher`, which defaults to False, rather than True.
(`165 <https://github.com/HazyResearch/fonduer/pull/165>`_)

Fixed
^^^^^
* `lukehsiao`_: Fixes the behavior of the ``get_between_ngrams`` data model
util. (`164 <https://github.com/HazyResearch/fonduer/pull/164>`_)
* `lukehsiao`_: Batch queries so that PostgreSQL buffers aren't exceeded.
(`162 <https://github.com/HazyResearch/fonduer/pull/162>`_)

-------------------

Changed
^^^^^^^
* `lukehsiao`_: :class:`MentionNgrams` ``split_tokens`` now defaults to an
empty list and splits on all occurrences, rather than just the first
occurrence.
* `j-rausch`_: Parser will now skip documents with parsing errors rather than
crashing.

Fixed
^^^^^
* `lukehsiao`_: Fix attribute error when using MentionFigures.

-------------------

Fixed
^^^^^
* `lukehsiao`_: Fix the layers module in fonduer.learning.disc_models.layers.