Alex Movila

odqa example gives RuntimeError in Win10

2

I run this code: from deeppavlov import configs from deeppavlov.core.common.file import read_json from deeppavlov import configs, train_model model_config = read_json(configs.doc_retrieval.en_ranker_tfidf_wiki) model_config["dataset_reader"]["data_path"] = "SentenceCorpus/unlabeled_articles/plos_unlabeled" model_config["dataset_reader"]["dataset_format"] = "txt" model_config["dataset_reader"]["save_path"] = "odqa/enwiki.db" model_config["dataset_iterator"]["load_path"]...

pls, tutorial on Medium?

1

Hi, Some tutorial on medium with use cases and explanation for dummies of these layers would be welcomed. It is not clear to me what are advantages of this library...

Bert model prediction is slow , consider more practical implementation

18

There is a new approach from AliBaba described here. [ALIBABA AI BEATS HUMANS IN READING-COMPREHENSION TEST](https://www.alizila.com/alibaba-ai-beats-humans-in-reading-comprehension-test-again) [A Deep Cascade Model for Multi-Document Reading Comprehension](https://arxiv.org/abs/1811.11374) It seems faster with better scalability....

Make code more flexible & publish interfaces to be able to implement your own components

1

Make the code more modular to be able to implement your own reader , retriever etc. Document the interfaces.

For non-answerable questions does not seem to work

4

Hi, I tested with your script with downloaded model Bert-on-Squad-V2.0 I asked questions like: When BERT was born? Who created Earth? The answers are nonsensical instead of answering "I can't...

How to avoid showing button payload text in UI?

9

Hi, After pressing button to reply the payload text is shown in the chat window. This happens in Chrome. Can I avoid this? I use window.chatroom = new window.Chatroom