# Updating an NER model using the annotation tool

**URL:** <https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578>\
**Category:** Uncategorized\
**Tags:** ner, spacy\
**Created:** [June 2, 2023, 1:19am UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578 "2023-06-02T01:19:30Z")\
**Posts on this page:** 7\
**Page:** 1

<div class="post-metadata">

**Author:** ![ModeEdna](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/modeedna/32/3821_2.png) [@ModeEdna](https://support.prodi.gy/u/ModeEdna)\
**Post date:** [June 2, 2023, 1:19am UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/1 "2023-06-02T01:19:30Z")

</div>

I'm interested in grabbing one of the pre-existing language models from Space and retraining it to better perform NER tasks on the texts I'm using. After going through documentation and watching some videos, I'm still struggling to understand how to go through the whole process. For example: Which format should the data be in (csv, jsonl, etc.)? Should the data just be the text, or should it include other information? And so on.

Is there an article or video that I've missed that has a step-by-step example of training an NER model? For context, I've watched the video on training an NER model for ingredient tags and also gone through a good chunk of the NER documentation.

Any help is appreciated!

---

<div class="post-metadata">

**Author:** ![koaning](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/koaning/32/230_2.png) [@koaning](https://support.prodi.gy/u/koaning)\
**Post date:** [June 2, 2023, 9:21am UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/2 "2023-06-02T09:21:15Z")

</div>

> [@ModeEdna](#):
>
> I'm interested in grabbing one of the pre-existing language models from Space and retraining it to better perform NER tasks on the texts I'm using.

Have you read the getting started guide mentioned below?

> **[Prodigy 101 – everything you need to know · Prodigy · An annotation tool for...](https://prodi.gy/docs/)**
>
> A downloadable annotation tool for NLP and computer vision tasks such as named entity recognition, text classification, object detection, image segmentation, A/B evaluation and more.

In this guide, you'll be using the `blank:en` spaCy pipeline, which is a pipeline with just a tokeniser to annotate some examples. It's this command:

 ![CleanShot 2023-06-02 at 11.16.54](https://us1.discourse-cdn.com/flex020/uploads/prodigy/original/2X/b/b8b355f4a77477a1eea8bdcea884b7be22220534.png)

Then, after you've added some annotations, you'll notice that the same guides will have you run a train command. In this section:

 ![CleanShot 2023-06-02 at 11.17.32](https://us1.discourse-cdn.com/flex020/uploads/prodigy/original/2X/3/3677743d00ae487c3475edbb6b32048b5fd4df06.png)

The thing with that train command is that you're allmost there, you can check the documentation for the train recipe to check some of the settings.

> **[Built-in Recipes · Prodigy · An annotation tool for AI, Machine Learning...](https://prodi.gy/docs/recipes/#training)**
>
> A downloadable annotation tool for NLP and computer vision tasks such as named entity recognition, text classification, object detection, image segmentation, A/B evaluation and more.

In particular, notice the `--base-model` setting. This allows you to pass in an existing pipeline as a starting point. So effectively, you'd need to change the train command from this:

```python
python -m prodigy train --ner new_news_headlines

```

Into this:

```python
python -m prodigy train --ner new_news_headlines --base-model en_core_web_md trained_model

```

This will now create a `trained_model` folder, which contains a spacy pipeline that you can load via:

```python
import spacy 

nlp = spacy.load("trained_model")

```

Does this help?

---

<div class="post-metadata">

**Author:** ![ModeEdna](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/modeedna/32/3821_2.png) [@ModeEdna](https://support.prodi.gy/u/ModeEdna)\
**Post date:** [June 2, 2023, 1:14pm UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/3 "2023-06-02T13:14:22Z")

</div>

Hi Vincent, thanks for the response! Yes, I did go over all that documentation, but the way you lay it out makes it even easier to understand. I guess my main confusion is with the data. Can the data come from a csv file on my computer, or does it have to be in another format? I noticed the examples use jsonl, but I have more experience using csv, so I was wondering if it's possible to use it. Also curious about the data itself... can it be raw text or do I have to do any preprocessing on it? Thanks again!

---

<div class="post-metadata">

**Author:** ![koaning](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/koaning/32/230_2.png) [@koaning](https://support.prodi.gy/u/koaning)\
**Post date:** [June 2, 2023, 1:28pm UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/4 "2023-06-02T13:28:34Z")

</div>

The details on file formats can also be found on our docs here:

> **[Loaders and Input Data · Prodigy · An annotation tool for AI, Machine...](https://prodi.gy/docs/api-loaders/)**
>
> A downloadable annotation tool for NLP and computer vision tasks such as named entity recognition, text classification, object detection, image segmentation, A/B evaluation and more.

That said, CSVs are fine just but JSONL files are somewhat preferred. The main thing that's important is that you have the right keys/column names. The `ner.manual` recipe, for example, needs to have a "text" column around in order to render the annotation card.

All the recipes that Prodigy provide will handle any of the required preprocessing, it'll only become a concern once you start writing custom recipes. In that case though, you can check this section of the docs:

> **[Annotation interfaces · Prodigy · An annotation tool for AI, Machine Learning...](https://prodi.gy/docs/api-interfaces/)**
>
> A downloadable annotation tool for NLP and computer vision tasks such as named entity recognition, text classification, object detection, image segmentation, A/B evaluation and more.

For each `view_id` that is used in a custom recipe, you can see what the expectations are in terms of input format.

 ![CleanShot 2023-06-02 at 15.28.01](https://us1.discourse-cdn.com/flex020/uploads/prodigy/original/2X/4/47ea02cd430146e6138e50e1fa13f49b01181da6.jpeg)

Is this sufficient? Feel free to ask for more details.

---

<div class="post-metadata">

**Author:** ![ModeEdna](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/modeedna/32/3821_2.png) [@ModeEdna](https://support.prodi.gy/u/ModeEdna)\
**Post date:** [June 5, 2023, 3:20pm UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/5 "2023-06-05T15:20:20Z")

</div>

Sorry, I still haven't managed to get it to work. I'm following the NER Manual recipe, and I'm inputting the following to start the annotations:

prodigy ner.manual es\_core\_news\_sm /Users/modeedna/Desktop/INEGI/ProyectoNER/output\_unescaped.jsonl --label LOC

I'm getting an error indicating that the source argument is required, which I thought was supposed to be the jsonl file with the text I want to annotate. Is the source meant to be something else? Also, if we don't select a database to store the annotations, it should be saving it to the .prodigy folder in the home directory, correct?

Again, thanks for the patience and help!

---

<div class="post-metadata">

**Author:** ![ryanwesslen](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/ryanwesslen/32/2969_2.png) [@ryanwesslen](https://support.prodi.gy/u/ryanwesslen)\
**Post date:** [June 5, 2023, 3:32pm UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/6 "2023-06-05T15:32:53Z")

</div>

hi @ModeEdna!

> [@ModeEdna](#):
>
> prodigy ner.manual es\_core\_news\_sm /Users/modeedna/Desktop/INEGI/ProyectoNER/output\_unescaped.jsonl --label LOC
> 
> I'm getting an error indicating that the source argument is required, which I thought was supposed to be the jsonl file with the text I want to annotate. Is the source meant to be something else?

You forgot to provide a name for your `dataset` that you'd annotate to, e.g., `ner_data`. It's a positional argument that follows right after `ner.manual`.

```terminal
$ python -m prodigy ner.manual --help
usage: prodigy ner.manual [-h] [-lo None] [-l None] [-pt None] [-e None] [-C] dataset spacy_model source

    Mark spans by token. Requires only a tokenizer and no entity recognizer,
    and doesn't do any active learning. If patterns are provided, their matches
    are highlighted in the example, if available. The recipe will present
    all examples in order, so even examples without matches are shown. If
    character highlighting is enabled, no "tokens" are saved to the database.
    

positional arguments:
  dataset Dataset to save annotations to
  spacy_model Loadable spaCy pipeline for tokenization or blank:lang (e.g. blank:en)
  source Data to annotate (file path or '-' to read from standard input)

```

So try:

```python
prodigy ner.manual ner_data es_core_news_sm /Users/modeedna/Desktop/INEGI/ProyectoNER/output_unescaped.jsonl --label LOC

```

> [@ModeEdna](#):
>
> Also, if we don't select a database to store the annotations, it should be saving it to the .prodigy folder in the home directory, correct?

Correct! By default, Prodigy sets the`PRODIGY_HOME` to `~/.prodigy` folder, which is where the `prodigy.json` (global) config file is (if you want to [manually configure Prodigy](https://prodi.gy/docs/install/#config)) and the `prodigy.db`, which is a SQLite DB by default.

You can confirm this location by running `prodigy stats` and viewing the `Prodigy Home` location.

---

<div class="post-metadata">

**Author:** ![ModeEdna](https://sea2.discourse-cdn.com/flex020/user_avatar/support.prodi.gy/modeedna/32/3821_2.png) [@ModeEdna](https://support.prodi.gy/u/ModeEdna)\
**Post date:** [June 5, 2023, 3:36pm UTC](https://support.prodi.gy/t/updating-an-ner-model-using-the-annotation-tool/6578/7 "2023-06-05T15:36:52Z")

</div>

Awesome, it's working now! Thanks again! I really appreciate the quick replies 🙂