Merge pull request snipsco#483 from snipsco/release/0.11.1

Release 0.11.1
kcubias · Feb 6, 2018 · 690696f · 690696f
2 parents 3401e34 + b5fc14e
commit 690696f
Show file tree

Hide file tree

Showing 14 changed files with 1,093 additions and 18 deletions.
diff --git a/Jenkinsfile b/Jenkinsfile
@@ -30,7 +30,7 @@ def buildWheel() {
     executeInVirtualEnv(pythonPath, "venv", cmd)
 }
 
-def installAndTest(pythonPath, venvPath, includeIntegrationTest=false) {
+def installAndTest(pythonPath, venvPath, includeIntegrationTest=false, includeLintingTest=false) {
     stage('Checkout') {
         deleteDir()
         checkout scm
@@ -54,6 +54,13 @@ def installAndTest(pythonPath, venvPath, includeIntegrationTest=false) {
         python -m unittest discover
         """
 
+        if(includeLintingTest) {
+            sh """
+            . ${venvPath}/bin/activate
+            python -m unittest discover -p 'linting_test*.py'
+            """
+        }
+
         if(includeIntegrationTest && (branchName.startsWith("release/") || branchName.startsWith("hotfix/") || branchName == "master")) {
             sh """
             . ${venvPath}/bin/activate
@@ -71,7 +78,7 @@ node('macos') {
     def python35path = sh(returnStdout: true, script: 'which python3.5').trim()
     def python36path = sh(returnStdout: true, script: 'which python3.6').trim()
 
-    installAndTest(python36path, "venv36", true)
+    installAndTest(python36path, "venv36", true, true)
     installAndTest(python27path, "venv27")
     installAndTest(python34path, "venv34")
     installAndTest(python35path, "venv35")

diff --git a/MANIFEST.in b/MANIFEST.in
@@ -0,0 +1,10 @@
+include snips_nlu/__version__
+recursive-include cli *.py
+recursive-include samples *
+recursive-include snips_nlu/snips-nlu-resources/de *
+recursive-include snips_nlu/snips-nlu-resources/en *
+recursive-include snips_nlu/snips-nlu-resources/es *
+recursive-include snips_nlu/snips-nlu-resources/fr *
+recursive-include snips_nlu/snips-nlu-resources/ko *
+recursive-include snips_nlu/snips-nlu-resources/zh *
+global-exclude __pycache__ *.py[cod]
diff --git a/README.md b/README.md
@@ -56,7 +56,7 @@ Install the package in edition mode:
 ## API
 
 ### Data
-The input of the NLU engine training is a dataset which format can be found [here](https://github.com/snipsco/snips-nlu/blob/master/snips_nlu/tests/resources/sample_dataset.json)
+The input of the NLU engine training is a dataset which format can be found [here](https://github.com/snipsco/snips-nlu/blob/master/samples/sample_dataset.json)
 
 ### Code
 
@@ -65,13 +65,13 @@ import io
 import json
 from pprint import pprint
 
-from snips_nlu import SnipsNLUEngine, NLUEngineConfig
+from snips_nlu import SnipsNLUEngine
 
 ############## Initialization ##############
 
 # The nlu config is optional here
 with io.open("config.json") as f:
-    nlu_config = NLUEngineConfig.from_dict(json.load(f))
+    nlu_config = json.load(f)
 
 engine = SnipsNLUEngine(config=nlu_config)
 
@@ -124,8 +124,8 @@ trained_engine = SnipsNLUEngine.from_dict(trained_engine_dict)
 ### CLI
 
 ```bash
->>> train-engine /path/to/input/dataset.json /path/to/output/trained_engine.json
->>> engine-inference /path/to/output/trained_engine.json
+train-engine /path/to/input/dataset.json /path/to/output/trained_engine.json
+engine-inference /path/to/output/trained_engine.json
 ```
 
 ### Versioning

diff --git a/samples/__init__.py b/samples/__init__.py
diff --git a/samples/configs/config_de.json b/samples/configs/config_de.json
@@ -0,0 +1,190 @@
+{
+  "unit_name": "nlu_engine",
+  "intent_parsers_configs": [
+    {
+      "unit_name": "deterministic_intent_parser",
+      "max_queries": 50,
+      "max_entities": 200
+    },
+    {
+      "unit_name": "probabilistic_intent_parser",
+      "slot_filler_config": {
+        "unit_name": "crf_slot_filler",
+        "feature_factory_configs": [
+          {
+            "args": {
+              "common_words_gazetteer_name": "top_10000_words",
+              "use_stemming": true,
+              "n": 1
+            },
+            "factory_name": "ngram",
+            "offsets": [
+              -2,
+              -1,
+              0,
+              1,
+              2
+            ]
+          },
+          {
+            "args": {
+              "common_words_gazetteer_name": "top_10000_words",
+              "use_stemming": true,
+              "n": 2
+            },
+            "factory_name": "ngram",
+            "offsets": [
+              -2,
+              1
+            ]
+          },
+          {
+            "args": {
+              "prefix_size": 2
+            },
+            "factory_name": "prefix",
+            "offsets": [
+              0
+            ]
+          },
+          {
+            "args": {
+              "prefix_size": 5
+            },
+            "factory_name": "prefix",
+            "offsets": [
+              0
+            ]
+          },
+          {
+            "args": {
+              "suffix_size": 2
+            },
+            "factory_name": "suffix",
+            "offsets": [
+              0
+            ]
+          },
+          {
+            "args": {
+              "suffix_size": 5
+            },
+            "factory_name": "suffix",
+            "offsets": [
+              0
+            ]
+          },
+          {
+            "args": {},
+            "factory_name": "is_digit",
+            "offsets": [
+              -1,
+              0,
+              1
+            ]
+          },
+          {
+            "args": {},
+            "factory_name": "is_first",
+            "offsets": [
+              -2,
+              -1,
+              0
+            ]
+          },
+          {
+            "args": {},
+            "factory_name": "is_last",
+            "offsets": [
+              0,
+              1,
+              2
+            ]
+          },
+          {
+            "args": {
+              "n": 1
+            },
+            "factory_name": "shape_ngram",
+            "offsets": [
+              0
+            ]
+          },
+          {
+            "args": {
+              "n": 2
+            },
+            "factory_name": "shape_ngram",
+            "offsets": [
+              -1,
+              0
+            ]
+          },
+          {
+            "args": {
+              "n": 3
+            },
+            "factory_name": "shape_ngram",
+            "offsets": [
+              -1
+            ]
+          },
+          {
+            "args": {
+              "use_stemming": true,
+              "tagging_scheme_code": 2
+            },
+            "factory_name": "entity_match",
+            "offsets": [
+              -2,
+              -1,
+              0
+            ],
+            "drop_out": 0.5
+          },
+          {
+            "args": {
+              "tagging_scheme_code": 1
+            },
+            "factory_name": "builtin_entity_match",
+            "offsets": [
+              -2,
+              -1,
+              0
+            ]
+          }
+        ],
+        "crf_args": {
+          "c1": 0.1,
+          "c2": 0.1,
+          "algorithm": "lbfgs"
+        },
+        "tagging_scheme": 1,
+        "entities_offsets": [
+          -2,
+          -1,
+          0
+        ],
+        "exhaustive_permutations_threshold": 64,
+        "data_augmentation_config": {
+          "min_utterances": 200,
+          "capitalization_ratio": 0.2
+        },
+        "random_seed": null
+      },
+      "intent_classifier_config": {
+        "unit_name": "log_reg_intent_classifier",
+        "data_augmentation_config": {
+          "min_utterances": 20,
+          "noise_factor": 5,
+          "unknown_word_prob": 0.0,
+          "unknown_words_replacement_string": null
+        },
+        "featurizer_config": {
+          "sublinear_tf": false
+        },
+        "random_seed": null
+      }
+    }
+  ]
+}