Added hard lexical constraints (Post & Vilar, NAACL 2018) (#349)

* Added hard lexical constraints (Post & Vilar, NAACL 2018)

This commit adds hard lexical constraints as described in Post & Vilar, 
Fast Lexically Constrained Decoding with Dynamic Beam Allocation for Neural Machine Translation (NAACL 2018).
Hard lexical constraints are words or phrases that must appear in the translation output.

To use this feature, for a particular input sentences, create a (1-line) JSON object of the following form:
```
    { "text": "Einer soll ein hoch@@ rangi@@ ges Mitglied aus Berlin gewesen sein .",
      "constraints": ["is said to", "powerful"] }
```

You then need to pass the JSON input flag (`--json-input`) to `sockeye.translate`.

CHANGELOG.md

@@ -10,6 +10,10 @@ Note that Sockeye has checks in place to not translate with an old model that wa
 
 Each version section may have have subsections for: _Added_, _Changed_, _Removed_, _Deprecated_, and _Fixed_.
 
+## [1.18.15]
+### Added
+- Added decoding with target-side lexical constraints (documentation in `tutorials/constraints`).
+
 ## [1.18.14]
 ### Added
 - Introduced Sockeye Autopilot for single-command end-to-end system building.

docs/modules.rst

@@ -106,6 +106,13 @@ sockeye.layers module
     :members:
     :show-inheritance:
 
+sockeye.lexical_constraints module
+----------------------------------
+
+.. automodule:: sockeye.lexical_constraints
+    :members:
+    :show-inheritance:
+
 sockeye.lexicon module
 ----------------------
 

sockeye/__init__.py

@@ -11,4 +11,4 @@
 # express or implied. See the License for the specific language governing
 # permissions and limitations under the License.
 
-__version__ = '1.18.14'
+__version__ = '1.18.15'

sockeye/constants.py

@@ -192,8 +192,12 @@
 # Inference Input JSON constants
 JSON_TEXT_KEY = "text"
 JSON_FACTORS_KEY = "factors"
+JSON_CONSTRAINTS_KEY = "constraints"
 JSON_ENCODING = "utf-8"
 
+# Lexical constraints
+BANK_ADJUSTMENT = 'even'
+
 VERSION_NAME = "version"
 CONFIG_NAME = "config"
 LOG_NAME = "log"