Skip to content

Commit b86f14f

Browse files
authored
Merge pull request RasaHQ#5325 from RasaHQ/prepare-release-1.8.0
Prepare release 1.8.0
2 parents f7a66e9 + 65993c3 commit b86f14f

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

47 files changed

+429
-406
lines changed

.github/workflows/continous-integration.yml

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -69,7 +69,7 @@ jobs:
6969
with:
7070
path: ~/.cache/pypoetry/virtualenvs
7171
key: ${{ runner.os }}-poetry-3.7-${{ hashFiles('**/poetry.lock') }}
72-
restore-keys: ${{ runner.os }}-poetry-
72+
restore-keys: ${{ runner.os }}-poetry-3.7
7373

7474
- name: Install Dependencies 📦
7575
run: |
@@ -137,7 +137,7 @@ jobs:
137137
with:
138138
path: ~/.cache/pypoetry/virtualenvs
139139
key: ${{ runner.os }}-poetry-${{ matrix.python-version }}-${{ hashFiles('**/poetry.lock') }}
140-
restore-keys: ${{ runner.os }}-poetry-
140+
restore-keys: ${{ runner.os }}-poetry-${{ matrix.python-version }}
141141

142142
- name: Install Dependencies 📦
143143
run: |

.github/workflows/documentation.yml

Lines changed: 7 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -37,6 +37,13 @@ jobs:
3737
with:
3838
poetry-version: ${{ env.POETRY_VERSION }}
3939

40+
- name: Load Poetry Cached Libraries ⬇
41+
uses: actions/cache@v1
42+
with:
43+
path: ~/.cache/pypoetry/virtualenvs
44+
key: ${{ runner.os }}-poetry-3.6-non-full-${{ hashFiles('**/poetry.lock') }}
45+
restore-keys: ${{ runner.os }}-poetry-3.6-non-full
46+
4047
- name: Install Dependencies 📦
4148
run: make install
4249

CHANGELOG.rst

Lines changed: 145 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -17,6 +17,151 @@ This project adheres to `Semantic Versioning`_ starting with version 1.0.
1717

1818
.. towncrier release notes start
1919
20+
[1.8.0] - 2020-02-26
21+
^^^^^^^^^^^^^^^^^^^^
22+
23+
Deprecations and Removals
24+
-------------------------
25+
- `#4991 <https://github.com/rasahq/rasa/issues/4991>`_: Removed ``Agent.continue_training`` and the ``dump_flattened_stories`` parameter
26+
from ``Agent.persist``.
27+
- `#5266 <https://github.com/rasahq/rasa/issues/5266>`_: Properties ``Component.provides`` and ``Component.requires`` are deprecated.
28+
Use ``Component.required_components()`` instead.
29+
30+
Features
31+
--------
32+
- `#2674 <https://github.com/rasahq/rasa/issues/2674>`_: Add default value ``__other__`` to ``values`` of a ``CategoricalSlot``.
33+
34+
All values not mentioned in the list of values of a ``CategoricalSlot``
35+
will be mapped to ``__other__`` for featurization.
36+
- `#4088 <https://github.com/rasahq/rasa/issues/4088>`_: Add story structure validation functionality (e.g. `rasa data validate stories --max-history 5`).
37+
- `#5065 <https://github.com/rasahq/rasa/issues/5065>`_: Add :ref:`LexicalSyntacticFeaturizer` to sparse featurizers.
38+
39+
``LexicalSyntacticFeaturizer`` does the same featurization as the ``CRFEntityExtractor``. We extracted the
40+
featurization into a separate component so that the features can be reused and featurization is independent from the
41+
entity extraction.
42+
- `#5187 <https://github.com/rasahq/rasa/issues/5187>`_: Integrate language models from HuggingFace's `Transformers <https://github.com/huggingface/transformers>`_ Library.
43+
44+
Add a new NLP component :ref:`HFTransformersNLP` which tokenizes and featurizes incoming messages using a specified
45+
pre-trained model with the Transformers library as the backend.
46+
Add :ref:`LanguageModelTokenizer` and :ref:`LanguageModelFeaturizer` which use the information from
47+
:ref:`HFTransformersNLP` and sets them correctly for message object.
48+
Language models currently supported: BERT, OpenAIGPT, GPT-2, XLNet, DistilBert, RoBERTa.
49+
- `#5225 <https://github.com/rasahq/rasa/issues/5225>`_: Added a new CLI command ``rasa export`` to publish tracker events from a persistent
50+
tracker store using an event broker. See :ref:`section_export`, :ref:`tracker-stores`
51+
and :ref:`event-brokers` for more details.
52+
- `#5230 <https://github.com/rasahq/rasa/issues/5230>`_: Refactor how GPU and CPU environments are configured for TensorFlow 2.0.
53+
54+
Please refer to the :ref:`documentation <tensorflow_usage>` to understand
55+
which environment variables to set in what scenarios. A couple of examples are shown below as well:
56+
57+
.. code-block:: python
58+
59+
# This specifies to use 1024 MB of memory from GPU with logical ID 0 and 2048 MB of memory from GPU with logical ID 1
60+
TF_GPU_MEMORY_ALLOC="0:1024, 1:2048"
61+
62+
# Specifies that at most 3 CPU threads can be used to parallelize multiple non-blocking operations
63+
TF_INTER_OP_PARALLELISM_THREADS="3"
64+
65+
# Specifies that at most 2 CPU threads can be used to parallelize a particular operation.
66+
TF_INTRA_OP_PARALLELISM_THREADS="2"
67+
68+
- `#5266 <https://github.com/rasahq/rasa/issues/5266>`_: Added a new NLU component :ref:`DIETClassifier <diet-classifier>` and a new policy :ref:`TEDPolicy <ted_policy>`.
69+
70+
DIET (Dual Intent and Entity Transformer) is a multi-task architecture for intent classification and entity
71+
recognition. You can read more about this component in our :ref:`documentation <diet-classifier>`.
72+
The new component will replace the :ref:`EmbeddingIntentClassifier <embedding-intent-classifier>` and the
73+
:ref:`CRFEntityExtractor` in the future.
74+
Those two components are deprecated from now on.
75+
See :ref:`migration guide <migration-to-rasa-1.8>` for details on how to
76+
switch to the new component.
77+
78+
:ref:`TEDPolicy <ted_policy>` is the new name for :ref:`EmbeddingPolicy <embedding_policy>`.
79+
``EmbeddingPolicy`` is deprecated from now on.
80+
The functionality of ``TEDPolicy`` and ``EmbeddingPolicy`` is the same.
81+
Please update your configuration file to use the new name for the policy.
82+
- `#663 <https://github.com/rasahq/rasa/issues/663>`_: The sentence vector of the ``SpacyFeaturizer`` and ``MitieFeaturizer`` can be calculated using max or mean pooling.
83+
84+
To specify the pooling operation, set the option ``pooling`` for the ``SpacyFeaturizer`` or the ``MitieFeaturizer``
85+
in your configuration file. The default pooling operation is ``mean``. The mean pooling operation also does not take
86+
into account words, that do not have a word vector.
87+
See our :ref:`documentation <components>` for more details.
88+
89+
Improvements
90+
------------
91+
- `#3975 <https://github.com/rasahq/rasa/issues/3975>`_: Added command line argument ``--conversation-id`` to ``rasa interactive``.
92+
If the argument is not given, ``conversation_id`` defaults to a random uuid.
93+
- `#4653 <https://github.com/rasahq/rasa/issues/4653>`_: Added a new command-line argument ``--init-dir`` to command ``rasa init`` to specify
94+
the directory in which the project is initialised.
95+
- `#4682 <https://github.com/rasahq/rasa/issues/4682>`_: Added support to send images with the twilio output channel.
96+
- `#4817 <https://github.com/rasahq/rasa/issues/4817>`_: Part of Slack sanitization:
97+
Multiple garbled URL's in a string coming from slack will be converted into actual strings.
98+
``Example: health check of <http://eemdb.net|eemdb.net> and <http://eemdb1.net|eemdb1.net> to health check of
99+
eemdb.net and eemdb1.net``
100+
- `#5117 <https://github.com/rasahq/rasa/issues/5117>`_: New command-line argument --conversation-id will be added and wiil give the ability to
101+
set specific conversation ID for each shell session, if not passed will be random.
102+
- `#5211 <https://github.com/rasahq/rasa/issues/5211>`_: Messages sent to the :ref:`event-brokers-pika` are now persisted. This guarantees
103+
the RabbitMQ will re-send previously received messages after a crash. Note that this
104+
does not help for the case where messages are sent to an unavailable RabbitMQ instance.
105+
- `#5250 <https://github.com/rasahq/rasa/issues/5250>`_: Added support for mattermost connector to use bot accounts.
106+
- `#5266 <https://github.com/rasahq/rasa/issues/5266>`_: We updated our code to TensorFlow 2.
107+
108+
We added a new docker image for ConveRT.
109+
The new images uses the following configuration
110+
111+
.. code-block:: yaml
112+
113+
language: "en"
114+
115+
pipeline:
116+
- name: ConveRTTokenizer
117+
- name: ConveRTFeaturizer
118+
- name: RegexFeaturizer
119+
- name: LexicalSyntacticFeaturizer
120+
- name: CountVectorsFeaturizer
121+
- name: CountVectorsFeaturizer
122+
analyzer: "char_wb"
123+
min_ngram: 1
124+
max_ngram: 4
125+
- name: DIETClassifier
126+
- name: EntitySynonymMapper
127+
- name: ResponseSelector
128+
- `#5317 <https://github.com/rasahq/rasa/issues/5317>`_: Events exported using ``rasa export`` receive a message header if published through a
129+
``PikaEventBroker``. The header is added to the message's ``BasicProperties.headers``
130+
under the ``rasa-export-process-id`` key
131+
(``rasa.core.constants.RASA_EXPORT_PROCESS_ID_HEADER_NAME``). The value is a
132+
UUID4 generated at each call of ``rasa export``. The resulting header is a key-value
133+
pair that looks as follows:
134+
135+
.. code-block:: text
136+
137+
'rasa-export-process-id': 'd3b3d3ffe2bd4f379ccf21214ccfb261'
138+
139+
- `#5292 <https://github.com/rasahq/rasa/issues/5292>`_: Added ``followlinks=True`` to os.walk calls, to allow the use of symlinks in training, NLU and domain data.
140+
- `#4811 <https://github.com/rasahq/rasa/issues/4811>`_: Support invoking a ``SlackBot`` by direct messaging or ``@<app name>`` mentions.
141+
142+
Bugfixes
143+
--------
144+
- `#4006 <https://github.com/rasahq/rasa/issues/4006>`_: Fixed timestamp parsing warning when using DucklingHTTPExtractor
145+
- `#4601 <https://github.com/rasahq/rasa/issues/4601>`_: Fixed issue with ``action_restart`` getting overridden by ``action_listen`` when the ``MappingPolicy`` and the
146+
`TwoStageFallbackPolicy <https://rasa.com/docs/rasa/core/policies/#two-stage-fallback-policy>`_ are used together.
147+
- `#5201 <https://github.com/rasahq/rasa/issues/5201>`_: Fixed incorrectly raised Error encountered in pipelines with a ``ResponseSelector`` and NLG.
148+
149+
When NLU training data is split before NLU pipeline comparison,
150+
NLG responses were not also persisted and therefore training for a pipeline including the ``ResponseSelector`` would fail.
151+
152+
NLG responses are now persisted along with NLU data to a ``/train`` directory in the ``run_x/xx%_exclusion`` folder.
153+
- `#5277 <https://github.com/rasahq/rasa/issues/5277>`_: Fixed sending custom json with Twilio channel
154+
155+
Improved Documentation
156+
----------------------
157+
- `#5174 <https://github.com/rasahq/rasa/issues/5174>`_: Updated the documentation to properly suggest not to explicitly add utterance actions to the domain.
158+
- `#5189 <https://github.com/rasahq/rasa/issues/5189>`_: Added user guide for reminders and external events, including ``reminderbot`` demo.
159+
160+
Miscellaneous internal changes
161+
------------------------------
162+
- #3923, #4597, #4903, #5180, #5189, #5266, #699
163+
164+
20165
[1.7.4] - 2020-02-24
21166
^^^^^^^^^^^^^^^^^^^^
22167

changelog/2674.feature.rst

Lines changed: 0 additions & 4 deletions
This file was deleted.

changelog/3923.misc.rst

Lines changed: 0 additions & 3 deletions
This file was deleted.

changelog/3975.improvement.rst

Lines changed: 0 additions & 2 deletions
This file was deleted.

changelog/4006.bugfix.rst

Lines changed: 0 additions & 1 deletion
This file was deleted.

changelog/4088.feature.rst

Lines changed: 0 additions & 1 deletion
This file was deleted.

changelog/4597.misc.rst

Lines changed: 0 additions & 15 deletions
This file was deleted.

changelog/4601.bugfix.rst

Lines changed: 0 additions & 2 deletions
This file was deleted.

0 commit comments

Comments
 (0)