# SpaCy3 models evaluation on a custom dataset
**URL:** https://support.prodi.gy/t/spacy3-models-evaluation-on-a-custom-dataset/4413
**Category:** Uncategorized
**Tags:** solved, usage, training, spacy
**Created:** [July 6, 2021, 12:48am UTC](https://support.prodi.gy/t/spacy3-models-evaluation-on-a-custom-dataset/4413 "2021-07-06T00:48:35Z")
**Posts on this page:** 4
**Page:** 1
### Author:  [@LubaTovbin](https://support.prodi.gy/u/LubaTovbin)
#### Post date: [July 6, 2021, 12:48am UTC](https://support.prodi.gy/t/spacy3-models-evaluation-on-a-custom-dataset/4413/1 "2021-07-06T00:48:35Z")
Hi guys,
I have three custom datasets for the Parser, Tagger, and NER that were generated using Prodigy. Now, I want to use them to evaluate Parser, Tagger, and NER in the en\_core\_web\_trf model. Is there a way to do it without using Prodigy? The data format in my sets looks very different from the one described in SpaCy's Example class.
Any guidance is appreciated,
Thank you
---
### Author:  [@ines](https://support.prodi.gy/u/ines)
#### Post date: [July 6, 2021, 1:26am UTC](https://support.prodi.gy/t/spacy3-models-evaluation-on-a-custom-dataset/4413/2 "2021-07-06T01:26:04Z")
Hi! You can use the `data-to-spacy` recipe to export your annotations, which in Prodigy v1.10 will give you a corpus in spaCy's v2's JSON format. If you're using spaCy v3, you can run `spacy convert` to convert it to the binary format used by `spacy train`: [https://spacy.io/api/cli#convert](https://spacy.io/api/cli#convert) You'll then be able to [train and evaluate](https://spacy.io/usage/training) your model using a transformer-based config.
Btw, under the hood, spaCy v3's binary format is just a collection of annotated `Doc` objects (which now also makes it much easier to generate it programmatically): [https://spacy.io/api/data-formats#binary-training](https://spacy.io/api/data-formats#binary-training)
The upcoming Prodigy v1.11, currently available as a [nightly pre-release](https://support.prodi.gy/t/prodigy-nightly-spacy-v3-support-ui-for-overlapping-spans-improved-feeds-more/3861), will allow you to export your data in spaCy's `.spacy` format out-of-the-box.
---
### Author:  [@LubaTovbin](https://support.prodi.gy/u/LubaTovbin)
#### Post date: [July 6, 2021, 8:43pm UTC](https://support.prodi.gy/t/spacy3-models-evaluation-on-a-custom-dataset/4413/3 "2021-07-06T20:43:08Z")
Hi Ines,
Thanks for the response!
My data is in JSONL format. The "spacy convert" doesn't work on JSONL, am I wrong?
---
### Author:  [@ines](https://support.prodi.gy/u/ines)
#### Post date: [July 7, 2021, 1:13am UTC](https://support.prodi.gy/t/spacy3-models-evaluation-on-a-custom-dataset/4413/4 "2021-07-07T01:13:06Z")
Yes, the `spacy convert` command expects data in spaCy's JSON format for training, not just the raw annotations you've exported with Prodigy. You can create it using the `data-to-spacy` command, which will merge your annotations from different datasets and export a corpus in spaCy's format.