Show HN: Languagecrunch – NLP server Docker image

Nice!<p>Relevant links for anyone interested:<p>* spaCy on Github: <a href="https:&#x2F;&#x2F;github.com&#x2F;explosion&#x2F;spacy" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;explosion&#x2F;spacy</a><p>* NER demo: <a href="https:&#x2F;&#x2F;demos.explosion.ai&#x2F;displacy-ent&#x2F;" rel="nofollow">https:&#x2F;&#x2F;demos.explosion.ai&#x2F;displacy-ent&#x2F;</a><p>* Neural coref by HuggingFace: <a href="https:&#x2F;&#x2F;huggingface.co&#x2F;coref&#x2F;" rel="nofollow">https:&#x2F;&#x2F;huggingface.co&#x2F;coref&#x2F;</a><p>* Accuracy of built-in spaCy models: <a href="https:&#x2F;&#x2F;spacy.io&#x2F;usage&#x2F;facts-figures" rel="nofollow">https:&#x2F;&#x2F;spacy.io&#x2F;usage&#x2F;facts-figures</a><p>Last time I calculated, the lowest cost way to run spaCy in the cloud was on Google Compute Engine ns1-standard pre-emptable instances. It should be over 100x cheaper per document than using Google, Amazon or Microsoft&#x27;s cloud NLP APIs. Accuracy will depend on your problem, but if you have your own training data, performance should be similar.

What do you need on the docker host machine to run this? Any specific docker version? GPU?<p>Also, it would be useful to see the Dockerfile or script that generated this img.

Seems interesting! How can it be used with other languages than English?

The demo examples are wrong or don&#x27;t make much sense.<p>&quot;Donald Trump&#x27;s administration&quot; is not a person.<p>In the following example, &quot;The currency&quot; is not a subject and &quot;India&quot; is not an object.<p>I don&#x27;t know how much useful information is extracted by this system.

Source on github: <a href="https:&#x2F;&#x2F;github.com&#x2F;artpar&#x2F;languagecrunch" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;artpar&#x2F;languagecrunch</a>

@artpar can you post some docs on the endpoints and how they should be used. I want to tie this into a speech to text system but i need more api info

Cool! What corpus was this trained on?