# Many issues when customizing starter pack pipeline

**URL:** <https://forum.rasa.com/t/many-issues-when-customizing-starter-pack-pipeline/1198>\
**Category:** Rasa Open Source\
**Created:** [September 18, 2018, 11:59am UTC](https://forum.rasa.com/t/many-issues-when-customizing-starter-pack-pipeline/1198 "2018-09-18T11:59:46Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Mika629](https://avatars.discourse-cdn.com/v4/letter/m/d2c977/32.png) [@Mika629](https://forum.rasa.com/u/Mika629)\
**Post date:** [September 18, 2018, 11:59am UTC](https://forum.rasa.com/t/many-issues-when-customizing-starter-pack-pipeline/1198/1 "2018-09-18T11:59:46Z")

</div>

Hi everyone,

I face many issues when I try to customize parameters of the Rasa pipeline. Starting from the [Rasa Starter Pack](https://github.com/RasaHQ/starter-pack-rasa-nlu), I first try the following pipeline (suggested [here](https://rasa.com/docs/nlu/choosing_pipeline/)) without issue:

`language: "en"`

`pipeline:`

`- name: "tokenizer_whitespace"`

`- name: "ner_crf"`

`- name: "intent_featurizer_count_vectors"`

`- name: "intent_classifier_tensorflow_embedding"`

Then, I try to customize this pipeline with configurations presented [here](https://rasa.com/docs/nlu/components/). First I wanted the customize the [intent\_classifier\_tensorflow\_embedding](https://rasa.com/docs/nlu/components/#intent-classifier-tensorflow-embedding), by copying/pasting his detailed configuration. In my opinion the detailed configuration is the default configuration, therefore I switch `- name: "intent_classifier_tensorflow_embedding"` with the complete pipeline and there is no issue as the configuration remains the same.

_1. Problems with `max_features` of `intent_featurizer_count_vectors`_

However, issues come when I do the same thing with `intent_featurizer_count_vectors` and `ner_crf`. When I switch only the `intent_featurizer_count_vectors` with its complete pipeline (described [here](https://rasa.com/docs/nlu/components/#intent-featurizer-count-vectors)), I have first the following issue with the `max_features` parameter which is considered as the string ‘None’ instead of a “real” None: `ValueError: max_features='None', neither a positive integer nor None` I can avoid this value by deleting the corresponding line, but it remains an issue when it is mentioned.

_2. Problems with features shapes with `intent_featurizer_count_vectors`_

Then, if I delete the `max_features` line, whatever the detailed pipeline I set up for `intent_featurizer_count_vectors`, I get the following issue with tensorflow classifier: `ValueError: Cannot feed value of shape (64,) for Tensor 'a:0', which has shape '(?, 66)'`

It seems that the `Tensor` is not reshaped, which explains the `?`.

**Therefore, my first question is: How can I solve this problem? How can I customize the “intent\_featurizer\_count\_vectors” when it is followed by the “intent\_classifier\_tensorflow\_embedding”?**

_3. Problems with `ner_crf`_

Next, I have also the same kind of problems with `ner_crf`. When I set only the default detailed pipeline described [here](https://rasa.com/docs/nlu/components/#ner-crf), I first get an error due to the features `pos` and `pos2` unavailable without loading `nlp_spacy` as described [here](https://rasa.com/docs/nlu/components/#nlp-spacy).

**So my second question is: How can `ner_crf` without the detailed configuration work if `nlp_spacy` is not in the pipeline? Does it mean that the configuration on the hyperlink is not the default configuration?**

Therefore, I loaded the `nlp_spacy` before the detailed `ner_crf` configuration. It solved the problem with my own data set, but I just see that it does not work with the starter pack because of a `index out of range` error.

You can try to reproduce these errors from the starter pack… If you find the solution that would help me and also the Rasa team to fix these issues with default configurations…

Thanks a lot in advance for your answers! 🙂

---

<div class="post-metadata">

**Author:** ![karandesaiii](https://avatars.discourse-cdn.com/v4/letter/k/85e7bf/32.png) [@karandesaiii](https://forum.rasa.com/u/karandesaiii)\
**Post date:** [November 20, 2018, 9:29am UTC](https://forum.rasa.com/t/many-issues-when-customizing-starter-pack-pipeline/1198/2 "2018-11-20T09:29:20Z")

</div>

@Abir

---

<div class="post-metadata">

**Author:** ![karandesaiii](https://avatars.discourse-cdn.com/v4/letter/k/85e7bf/32.png) [@karandesaiii](https://forum.rasa.com/u/karandesaiii)\
**Post date:** [November 20, 2018, 9:38am UTC](https://forum.rasa.com/t/many-issues-when-customizing-starter-pack-pipeline/1198/3 "2018-11-20T09:38:36Z")

</div>

@akelad Please help out with issue 1 and 2

---

<div class="post-metadata">

**Author:** ![Abir](https://dub1.discourse-cdn.com/flex013/user_avatar/forum.rasa.com/abir/32/771_2.png) [@Abir](https://forum.rasa.com/u/Abir)\
**Post date:** [November 29, 2018, 6:08am UTC](https://forum.rasa.com/t/many-issues-when-customizing-starter-pack-pipeline/1198/4 "2018-11-29T06:08:28Z")

</div>

> [@\[SOLVED\]ValueError: Cannot feed value of shape (64,) for Tensor 'a:0', which has shape '(?, 945)'](http://forum.rasa.com/t/solved-valueerror-cannot-feed-value-of-shape-64-for-tensor-a-0-which-has-shape-945/2403):
>
> I used the below config but as i train the model an error is thrown language : “en” pipeline: name: “tokenizer\_whitespace” name: “ner\_crf” name: “intent\_featurizer\_count\_vectors” “token\_pattern”: r’(?u)\b\w+\b’ name: “intent\_classifier\_tensorflow\_embedding” am i doing something wrong ? @akelad

Go through this link and see if you made the same mistake . If not then share the complete nlu\_config
