# Will a GPU make training faster?

**URL:** <https://support.prodi.gy/t/will-a-gpu-make-training-faster/187>\
**Category:** Uncategorized\
**Tags:** spacy\
**Created:** [January 4, 2018, 7:50pm UTC](https://support.prodi.gy/t/will-a-gpu-make-training-faster/187 "2018-01-04T19:50:03Z")\
**Posts on this page:** 1\
**Showing post:** 5

<div class="post-metadata">

**Author:** ![ines](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ines/32/3_2.png) [@ines](https://support.prodi.gy/u/ines)\
**Post date:** [February 2, 2018, 1:54am UTC](https://support.prodi.gy/t/will-a-gpu-make-training-faster/187/5 "2018-02-02T01:54:33Z")

</div>

> [@SandeepNaidu](#):
>
> How can I use “spacy train” instead of “prodigy batch-train” with prodigy’s annotations? I think I missed some reading regarding this.

[Here's a thread on this topic](https://support.prodi.gy/t/prodigy-annotations-to-spacy-train/243), including example scripts. Version 1.3.0 of Prodigy, which we've just released includes a new [`ner.gold-to-spacy` converter recipe](https://prodi.gy/docs/recipes#ner-gold-to-spacy) that exports training data in spaCy's formats (either the "simple training style" or BILUO scheme annotations). You can also check out the source of that recipe for inspiration, and write your own conversion script.

---

_[View the full topic](https://support.prodi.gy/t/will-a-gpu-make-training-faster/187)._