# Ner\_crf

**URL:** <https://forum.rasa.com/t/ner-crf/769>\
**Category:** Rasa Open Source\
**Created:** [August 30, 2018, 2:30pm UTC](https://forum.rasa.com/t/ner-crf/769 "2018-08-30T14:30:08Z")\
**Posts on this page:** 13\
**Page:** 1

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [August 30, 2018, 2:30pm UTC](https://forum.rasa.com/t/ner-crf/769/1 "2018-08-30T14:30:08Z")

</div>

Hey, I am facing a little ner\_crf, it detects new as entity in both sentences like, I am for a new website, I am not looking at new website, it must recognized new in second sentence .

here it my pipeline:

- name: “nlp\_spacy” model: “en”
- name: “tokenizer\_spacy”
- name: “ner\_crf” BILOU\_flag: true features:

# features for word before token

- [“low”, “title”, “upper”, “pos”, “pos2”]

# features of token itself

- [“bias”, “low”, “upper”, “title”, “digit”, “pos”, “pos2”, “pattern”]

# features for word after the token we want to tag

- [“low”, “title”, “upper”, “pos”, “pos2”] max\_iterations: 50 L1\_c: 1 L2\_c: 1e-3
- name: “ner\_synonyms”
- name: “intent\_featurizer\_count\_vectors”
- name: “intent\_classifier\_tensorflow\_embedding” ![image](https://europe1.discourse-cdn.com/flex013/uploads/rasa/original/1X/f4fd2f34ee6f00ebbcc6af1b1d25fca37bd24232.png)

and I face when I add “word2” , “word3” in crf feature in the pipline I face this error

---

<div class="post-metadata">

**Author:** ![akelad](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/akelad/32/69_2.png) [@akelad](https://forum.rasa.com/u/akelad)\
**Post date:** [September 3, 2018, 8:03am UTC](https://forum.rasa.com/t/ner-crf/769/2 "2018-09-03T08:03:42Z")

</div>

We made some changes to the features in ner\_crf, this is the list of available ones now:

> <https://github.com/RasaHQ/rasa_nlu/blob/master/rasa_nlu/extractors/crf_entity_extractor.py#L67>

The standard ones used are listed here: [http://rasa.com/docs/nlu/components/#ner-crf](http://rasa.com/docs/nlu/components/#ner-crf)

---

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [September 3, 2018, 8:47am UTC](https://forum.rasa.com/t/ner-crf/769/3 "2018-09-03T08:47:33Z")

</div>

I have used the latest version of rasa\_nlu and I have also updated the dictionary but all in vain, I am facing the same issue.

---

<div class="post-metadata">

**Author:** ![akelad](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/akelad/32/69_2.png) [@akelad](https://forum.rasa.com/u/akelad)\
**Post date:** [September 4, 2018, 9:30am UTC](https://forum.rasa.com/t/ner-crf/769/4 "2018-09-04T09:30:11Z")

</div>

Updated the dictionary where? Can you post your config file in a legible format please?

---

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [September 4, 2018, 9:38am UTC](https://forum.rasa.com/t/ner-crf/769/5 "2018-09-04T09:38:38Z")

</div>

```auto
language: "en"

pipeline:
- name: "nlp_spacy"
  model: "en"
- name: "tokenizer_spacy"
- name: "ner_spacy"
- name: "ner_duckling"
- name: "ner_duckling_http"
  dimensions:
  - "NUMBER"
- name: "ner_crf"
  BILOU_flag: true
  features:
    # features for word before token
    - ["low", "title", "upper", "pos", "pos2"]
    # features of token itself
    - ["bias", "low", "upper","word3", "word2" "title", "digit", "pos", "pos2", "pattern"]
    # features for word after the token we want to tag
    - ["low", "title", "upper", "pos", "pos2"]
  max_iterations: 50
  L1_c: 1
  L2_c: 1e-3
- name: "ner_synonyms"
- name: "intent_featurizer_count_vectors"
- name: "intent_classifier_tensorflow_embedding"
```

---

<div class="post-metadata">

**Author:** ![akelad](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/akelad/32/69_2.png) [@akelad](https://forum.rasa.com/u/akelad)\
**Post date:** [September 5, 2018, 2:38pm UTC](https://forum.rasa.com/t/ner-crf/769/6 "2018-09-05T14:38:14Z")

</div>

Yes so you need to remove all the `word3`, `word2` etc features that aren’t part of the available features for the new version anymore. These are the standard ones, that I linked previously [http://rasa.com/docs/nlu/components/#ner-crf](http://rasa.com/docs/nlu/components/#ner-crf)

---

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [September 6, 2018, 12:27pm UTC](https://forum.rasa.com/t/ner-crf/769/7 "2018-09-06T12:27:20Z")

</div>

Can you tell why it is not update in git ?

> <https://github.com/RasaHQ/rasa_nlu/blob/master/sample_configs/config_crf.yml>

---

<div class="post-metadata">

**Author:** ![akelad](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/akelad/32/69_2.png) [@akelad](https://forum.rasa.com/u/akelad)\
**Post date:** [September 6, 2018, 12:37pm UTC](https://forum.rasa.com/t/ner-crf/769/8 "2018-09-06T12:37:41Z")

</div>

looks like we forgot to update that, feel free to create a PR to fix it

---

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [September 6, 2018, 12:44pm UTC](https://forum.rasa.com/t/ner-crf/769/9 "2018-09-06T12:44:05Z")

</div>

Ok thanks

---

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [September 6, 2018, 12:46pm UTC](https://forum.rasa.com/t/ner-crf/769/10 "2018-09-06T12:46:50Z")

</div>

Can you tell where I can where I can find help, I would like to know what this features means, like “pos”, “title” etc. as mentioned in spacy config [http://rasa.com/docs/nlu/components/#ner-crf](http://rasa.com/docs/nlu/components/#ner-crf)

---

<div class="post-metadata">

**Author:** ![azizullah2017](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/azizullah2017/32/241_2.png) [@azizullah2017](https://forum.rasa.com/u/azizullah2017)\
**Post date:** [September 6, 2018, 1:04pm UTC](https://forum.rasa.com/t/ner-crf/769/11 "2018-09-06T13:04:07Z")

</div>

@akelad can you help where to find the features meaning ?

---

<div class="post-metadata">

**Author:** ![souvikg10](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/souvikg10/32/93_2.png) [@souvikg10](https://forum.rasa.com/u/souvikg10)\
**Post date:** [September 6, 2018, 2:21pm UTC](https://forum.rasa.com/t/ner-crf/769/12 "2018-09-06T14:21:29Z")

</div>

[https://eli5.readthedocs.io/en/latest/tutorials/sklearn\_crfsuite.html#feature-extraction](https://eli5.readthedocs.io/en/latest/tutorials/sklearn_crfsuite.html#feature-extraction)

You have information about the features here

---

<div class="post-metadata">

**Author:** ![kapilkathuria](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/kapilkathuria/32/766_2.png) [@kapilkathuria](https://forum.rasa.com/u/kapilkathuria)\
**Post date:** [September 28, 2018, 4:56am UTC](https://forum.rasa.com/t/ner-crf/769/13 "2018-09-28T04:56:21Z")

</div>

More information on crf features:

[https://hk.saowen.com/a/c8fe0764b2e5d63ca38cc9867746b739c94eb2bec0b2f6b86d24fa2b01023a11](https://hk.saowen.com/a/c8fe0764b2e5d63ca38cc9867746b739c94eb2bec0b2f6b86d24fa2b01023a11)