> For the complete documentation index, see [llms.txt](https://smartpredict.gitbook.io/smartpredict-ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://smartpredict.gitbook.io/smartpredict-ai/getting-started/getting-started-part1-iris-classification-project/step-4-build-flowchart/assembling-the-remaining-components-of-our-model.md).

# Set up the flowchart

In this part, we are going to remove unrequired elements and complete the flowchart by adding the remaining components of our model: the processing pipeline and the initial dataset .

### 1.  Deleting the dataframe loader

As we already processed our dataset , the dataframe loader is not needed anymore so we may delete it. To delete a module, click on the module menu (the three dots on its right) >> Then on Delete. An alert box asks for confirmation>> Click on OK.<br>

![Deleting unused modules is easy.](https://1833277725-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-Lyc1OXsKqB2S62LxsOR%2F-Lyxq_lwzMX06LIGHPsx%2F-Lyxt2D-NCmqMmfYEGap%2FSmartPredict_DeletingDFL.gif?alt=media\&token=1b922e6f-673d-4dc2-ac15-cacc1d3691b1)

### 2. Adding the processing pipeline and the dirty dataset

All other flowchart modules being already set into place, let us attach in the two last components .&#x20;

1\.  From the right sidebar, click on the *processing pipeline* icon. A list of processing pipelines shows. Look for the one we have just made earlier and drag and drop it into the workflow .                                                                                &#x20;

2\.  Then , on top of this latter, let us place the *dirty dataset* we initially uploaded (remember the iris dirty dataset?)&#x20;

Getting back to the main menu, click on the *Dashboard*  icon . Then from the right panel, find and click the *Dataset* icon (second from the left).  Look for our initial dataset iris\_dataset\_dirty.&#x20;

Once found,  connect this initial dataset on top of the processing pipeline to apply the processing mechanism onto it.&#x20;

![](https://1833277725-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-Lyc1OXsKqB2S62LxsOR%2F-LyttuTJ5qr6MYCT1jgL%2F-Lytuq2bzLG4_pz9bVDT%2Fimage.png?alt=media\&token=40064771-f966-45fe-b408-2b0edebcaa63)

{% hint style="info" %}
As always, a **dashed line** shows the path for connecting the output of a module to the input of another, which  definitely makes the task effortless.
{% endhint %}

&#x20;3\. Finally, connect the whole set $$(processingpipeline+dirty dataset)$$ to the **feature selector**.

![Adding the pipeline and the initial dataset.](https://1833277725-files.gitbook.io/~/files/v0/b/gitbook-legacy-files/o/assets%2F-Lyc1OXsKqB2S62LxsOR%2F-Lyxw9byqqd0ZuTvsmZT%2F-LyxxrcNZYq8KNU8ItjZ%2FSmartPredict_AddingElements.gif?alt=media\&token=bf0d2e32-908c-404f-895a-949a36e4dd52)