Reputation: 5095
I would like to know if there is a way in Google Colab that can collate outputs nicely, just like Markdown in R and how IPython Notebook can be converted to pdf and html format?
My output consists of multiple tables, graphs etc. I would like to preferably pretty print them into one file, of which some part are presentable enough to be used in a report.If there's no such method, what is the best alternative?
Upvotes: 3
Views: 30252
Reputation: 9
In Colab, an easy alternative will be to use the browser's print functionality.
file
on the toolbar & select print.Save as PDF
- this will download the notebook to your PC as pdf.Upvotes: 0
Reputation: 2576
You can also create a pdf in colab itself using nbconvert.
!apt update
!apt install texlive-xetex texlive-fonts-recommended texlive-generic-recommended
import re, pathlib, shutil
# Get a list of all your Notebooks
notebooks = [x for x in pathlib.Path("/content/drive/My Drive/Colab Notebooks").iterdir() if
re.search(r"\.ipynb", x.name, flags = re.I)]
for i, n in enumerate(notebooks):
print(f"\nProcessing [{i+1:{len(str(len(notebooks)))}d}/{len(notebooks)}] {n.name}\n")
# Optionally copy your notebooks from gdrive to your vm
shutil.copy(n, n.name)
n = pathlib.Path(n.name)
!jupyter nbconvert "{n.as_posix()}" --to pdf --output "{n.stem.replace(" ", "_")}"
Instead of using a magic to run nbconvert you could also use subprocess
s = subprocess.Popen(shlex.split(
f'jupyter nbconvert "{n.as_posix()}" --to pdf --output "{n.stem.replace(" ", "_")}"'
), shell = False, stdout = subprocess.PIPE, stderr = subprocess.PIPE)
s.wait()
s.stdout.read()
There are also more packages available regarding xetex in case you use a very sophisticated template.
sudo apt install pandoc nbconvert texlive texlive-latex-extra texlive-generic-extra
Upvotes: 5
Reputation: 26886
You can save / export an IPython notebook (menu: File
/ Download .ipynb
) and then use Jupyter to save to PDF.
Upvotes: 3