Nbconvert: Figures are left-aligned when converting --to html -no-input

[smerigon]

When converting a notebook to html and using the -no--input flag, figures (matplotlib, plotly) appear left-aligned, and hence are not aligned with the markdown cells anymore. Is it the intended behavior?
Using
jupyter-client==5.2.2
jupyter-console==5.2.0
jupyter-core==4.4.0
notebook==5.7.0
nbconvert==5.4.0

[simbalsyd]

I have a similar problem when converting to html. The left alignment is not the same when I define a cell as Markdown and when I write to Markdown from Python code. The converted document doesn't look nice.

I attached two files (alignment_test.ipynb and alignment_test.html). The html file was converted with:

jupyter nbconvert alignment_test.ipynb --no-input

Any idea how to get all text properly left-aligned in the html output?

It was not possible to attach ipynb and html file so I attached graphics instead

[fidelram]

I am having the same problem:

Without --no-input:

Here, the same notebook with --no-input (notice how the image is not aligned with the text):

After looking at the html code I noticed that a div element that gives the padding is missing in the --no-input case:

I hope this helps to solve the issue.

[xoelop]

Having the same problem here. Tried modified a bunch of templates but no success so far

Best solution I've found so far is this one:

--TemplateExporter.exclude_input=True

The problem is that in some cells, Out[number]: still appears:

EDIT
Using the flag --no-prompt kinda fixes it too. More info here

[th0ger]

Confirming @xoelop s workaround:
--to html --TemplateExporter.exclude_input=True --no-prompt
gives equivalent output, but with identical indentation for all content.

(Without the --no-prompt i similarly get Out[] for some cells. The Out[]'s seems randomly toggled thoughout the document.)

+1 for fixing the --no-input alignment.

[genzhupanda]

I faced the same problem but could not solve it. Hopefully, someone can help.

[MSeal]

Would love to have some more contributors help knock out the HTML issues that are open, I don't think any of the maintainers have had time to dig into these recently.