Features

Components

NER component

Based on neural Named Entity Recognition network. The NER component reproduces architecture from the paper Application of a Hybrid Bi-LSTM-CRF model to the task of Russian Named Entity Recognition which is inspired by Bi-LSTM+CRF architecture from https://arxiv.org/pdf/1603.01360.pdf.

Dataset Test F1
Persons-1000 dataset with additional LOC and ORG markup 95.25
DSTC 2 98.40
OntoNotes 87.07

Slot filling components

Based on fuzzy Levenshtein search to extract normalized slot values from text. The components either rely on NER results or perform needle in haystack search.

Dataset Slots Accuracy
DSTC 2 98.85

Classification component

Component for classification tasks (intents, sentiment, etc) on word-level. Shallow-and-wide CNN, Deep CNN, BiLSTM, BiLSTM with self-attention and other models are presented. The model also allows multilabel classification of texts. Several pre-trained models are available and presented in Table below.

Task Dataset Lang Model Metric Valid Test Downloads
28 intents DSTC 2 En DSTC 2 emb Accuracy 0.7732 0.7868 800 Mb
Wiki emb 0.9602 0.9593 8.5 Gb
7 intents SNIPS-2017 DSTC 2 emb F1 0.8685 800 Mb
Wiki emb 0.9811 8.5 Gb
Tfidf + SelectKBest + PCA + Wiki emb 0.9673 8.6 Gb
Wiki emb weighted by Tfidf 0.9786 8.5 Gb
Insult detection Insults Reddit emb ROC-AUC 0.9271 0.8618 6.2 Gb
5 topics AG News Wiki emb Accuracy 0.8876 0.9011 8.5 Gb
Sentiment Twitter mokoron Ru RuWiki+Lenta emb w/o preprocessing 0.9972 0.9971 6.2 Gb
RuWiki+Lenta emb with preprocessing 0.7811 0.7749 6.2 Gb
RuSentiment RuWiki+Lenta emb F1 0.6393 0.6539 6.2 Gb
ELMo 0.7066 0.7301 700 Mb
Intent Yahoo-L31 Yahoo-L31 on ELMo pre-trained on Yahoo-L6 ROC-AUC 0.9269 700 Mb

As no one had published intent recognition for DSTC-2 data, the comparison of the presented model is given on SNIPS dataset. The evaluation of model scores was conducted in the same way as in [3] to compare with the results from the report of the authors of the dataset. The results were achieved with tuning of parameters and embeddings trained on Reddit dataset.

Model AddToPlaylist BookRestaurant GetWheather PlayMusic RateBook SearchCreativeWork SearchScreeningEvent
api.ai 0.9931 0.9949 0.9935 0.9811 0.9992 0.9659 0.9801
ibm.watson 0.9931 0.9950 0.9950 0.9822 0.9996 0.9643 0.9750
microsoft.luis 0.9943 0.9935 0.9925 0.9815 0.9988 0.9620 0.9749
wit.ai 0.9877 0.9913 0.9921 0.9766 0.9977 0.9458 0.9673
snips.ai 0.9873 0.9921 0.9939 0.9729 0.9985 0.9455 0.9613
recast.ai 0.9894 0.9943 0.9910 0.9660 0.9981 0.9424 0.9539
amazon.lex 0.9930 0.9862 0.9825 0.9709 0.9981 0.9427 0.9581
               
Shallow-and-wide CNN 0.9956 0.9973 0.9968 0.9871 0.9998 0.9752 0.9854

Goal-oriented bot

Based on Hybrid Code Networks (HCNs) architecture from Jason D. Williams, Kavosh Asadi, Geoffrey Zweig, Hybrid Code Networks: practical and efficient end-to-end dialog control with supervised and reinforcement learning – 2017. It allows to predict responses in goal-oriented dialog. The model is customizable: embeddings, slot filler and intent classifier can be switched on and off on demand.

Available pre-trained models:

Dataset & Model Valid turn accuracy Test turn accuracy Downloads
DSTC2, bot with slot filler & intents 0.5288 0.5248 8.5 Gb
DSTC2, bot with slot filler & embeddings & attention 0.5538 0.5551 8.5 Gb

Other benchmarks on DSTC2 (can’t be directly compared due to dataset modifications):

Dataset & Model Test turn accuracy
DSTC2, Bordes and Weston (2016) 0.411
DSTC2, Perez and Liu (2016) 0.487
DSTC2, Eric and Manning (2017) 0.480
DSTC2, Williams et al. (2017) 0.556

Seq2seq goal-oriented bot

Dialogue agent predicts responses in a goal-oriented dialog and is able to handle multiple domains (pretrained bot allows calendar scheduling, weather information retrieval, and point-of-interest navigation). The model is end-to-end differentiable and does not need to explicitly model dialogue state or belief trackers.

Comparison of deeppavlov pretrained model with others:

Dataset & Model Valid BLEU Test BLEU Downloads
Kvret, KvretNet 0.1319 0.1328 10 Gb
Kvret, KvretNet, Mihail Eric et al. (2017) 0.132
Kvret, CopyNet, Mihail Eric et al. (2017) 0.110
Kvret, Attn Seq2Seq, Mihail Eric et al. (2017) 0.102
Kvret, Rule-based, Mihail Eric et al. (2017) 0.066

Automatic spelling correction component

Pipelines that use candidates search in a static dictionary and an ARPA language model to correct spelling errors.

Note

About 4.4 GB on disc required for the Russian language model and about 7 GB for the English one.

Comparison on the test set for the SpellRuEval competition on Automatic Spelling Correction for Russian:

Correction method Precision Recall F-measure Speed (sentences/s)
Yandex.Speller 83.09 59.86 69.59
Damerau Levenshtein 1 + lm 53.26 53.74 53.50 29.3
Brill Moore top 4 + lm 51.92 53.94 52.91 0.6
Hunspell + lm 41.03 48.89 44.61 2.1
JamSpell 44.57 35.69 39.64 136.2
Brill Moore top 1 41.29 37.26 39.17 2.4
Hunspell 30.30 34.02 32.06 20.3

Ranking component

Based on LSTM-based deep learning models for non-factoid answer selection. The model performs ranking of responses or contexts from some database by their relevance for the given context.

Available pre-trained models for ranking:

Dataset Model config Validation (Recall@1) Test1 (Recall@1) Downloads
InsuranceQA V1 ranking_insurance_interact 72.0 72.2 8374M
Ubuntu V2 ranking_ubuntu_v2_interact 52.9 52.4 8913M
Ubuntu V2 ranking_ubuntu_v2_mt_interact 59.2 58.7 8906M

Available pre-trained models for paraphrase identification:

Dataset Model config Val (accuracy) Test (accuracy) Val (F1) Test (F1) Val (log_loss) Test (log_loss) Downloads
paraphraser.ru paraphrase_ident_paraphraser 83.8 75.4 87.9 80.9 0.468 0.616 5938M
paraphraser.ru paraphrase_ident_paraphraser 82.7 76.0 87.3 81.4 0.391 0.510 5938M
paraphraser.ru paraphrase_ident_paraphraser_tune 82.9 76.7 87.3 82.0 0.392 0.479 5938M
Quora Question Pairs paraphrase_ident_qqp 87.1 87.0 83.0 82.6 0.300 0.305 8134M
Quora Question Pairs paraphrase_ident_qqp 87.7 87.5 84.0 83.8 0.287 0.298 8136M

Comparison with other models on the InsuranceQA V1:

Model Validation (Recall@1) Test1 (Recall@1)
Architecture II (HLQA(200) CNNQA(4000) 1-MaxPooling Tanh) 61.8 62.8
QA-LSTM basic-model(max pooling) 64.3 63.1
ranking_insurance 72.0 72.2

TF-IDF Ranker component

Based on Reading Wikipedia to Answer Open-Domain Questions. The model solves the task of document retrieval for a given query.

Dataset Model Wiki dump Recall@5 Downloads
SQuAD-v1.1 doc_retrieval enwiki (2018-02-11) 75.6 33 GB

Question Answering component

Based on R-NET: Machine Reading Comprehension with Self-matching Networks. The model solves the task of looking for an answer on a question in a given context (SQuAD task format).

Note

All pre-trained models could be downloaded. Model for English language will download about 2.5 Gb and model for Russian about 5 Gb.

Dataset Model config lang EM (dev) F-1 (dev)
SQuAD-v1.1 squad en 71.49 80.34
SDSJ Task B squad_ru ru 60.62 80.04

In the case when answer is not necessary present in given context we have squad_noans model. This model outputs empty string in case if there is no answer in context.

Morphological tagging component

Based on character-based approach to morphological tagging Heigold et al., 2017. An extensive empirical evaluation of character-based morphological tagging for 14 languages. A state-of-the-art model for Russian and several other languages. Model takes as input tokenized sentences and outputs the corresponding sequence of morphological labels in UD format. The table below contains word and sentence accuracy on UD2.0 datasets. For more scores see full table.

Dataset Model Word accuracy Sent. accuracy Download size (MB)
UD2.0 (Russian) Pymorphy + russian_tagsets (first tag) 60.93 0.00  
UD Pipe 1.2 (Straka et al., 2017) 93.57 43.04  
Basic model 95.17 50.58 48.7
Pymorphy-enhanced model 96.23 58.00 48.7
UD2.0 (Czech) UD Pipe 1.2 (Straka et al., 2017) 91.86 42.28  
Basic model 94.35 51.56 41.8
UD2.0 (English) UD Pipe 1.2 (Straka et al., 2017) 92.89 55.75  
Basic model 93.00 55.18 16.9
UD2.0 (German) UD Pipe 1.2 (Straka et al., 2017) 76.65 10.24  
Basic model 83.83 15.25 18.6

Frequently Asked Questions (FAQ) component

Set of pipelines for FAQ task: classifying incoming question into set of known questions and return prepared answer. You can build different pipelines based on: tf-idf, weighted fasttext, cosine similarity, logistic regression.

Skills

eCommerce bot

The eCommerce bot intends to retrieve product items from catalog in sorted order. In addition, it asks an user to provide additional information to specify the search.

Note

About 130 Mb on disc required for eCommerce bot with TfIdf-based ranker and 500 Mb for BLEU-based ranker.

ODQA

An open domain question answering skill. The skill accepts free-form questions about the world and outputs an answer based on its Wikipedia knowledge.

Dataset Model config Wiki dump F1 Downloads
SQuAD-v1.1 ODQA enwiki (2018-02-11) 28.0 42 GB

AutoML

Hyperparameters optimization

Hyperparameters optimization (either by cross-validation or neural evolution) for DeepPavlov models that requires only some small changes in a config file.

Embeddings

Pre-trained embeddings for the Russian language

Word vectors for the Russian language trained on joint Russian Wikipedia and Lenta.ru corpora.

Examples of some components

  • Run goal-oriented bot with Telegram interface:

    python -m deeppavlov interactbot deeppavlov/configs/go_bot/gobot_dstc2.json -d -t <TELEGRAM_TOKEN>

  • Run goal-oriented bot with console interface:

    python -m deeppavlov interact deeppavlov/configs/go_bot/gobot_dstc2.json -d

  • Run goal-oriented bot with REST API:

    python -m deeppavlov riseapi deeppavlov/configs/go_bot/gobot_dstc2.json -d

  • Run slot-filling model with Telegram interface:

    python -m deeppavlov interactbot deeppavlov/configs/ner/slotfill_dstc2.json -d -t <TELEGRAM_TOKEN>

  • Run slot-filling model with console interface:

    python -m deeppavlov interact deeppavlov/configs/ner/slotfill_dstc2.json -d

  • Run slot-filling model with REST API:

    python -m deeppavlov riseapi deeppavlov/configs/ner/slotfill_dstc2.json -d

  • Predict intents on every line in a file:

    python -m deeppavlov predict deeppavlov/configs/classifiers/intents_snips.json -d --batch-size 15 < /data/in.txt > /data/out.txt

View video demo of deployment of a goal-oriented bot and a slot-filling model with Telegram UI.