You need to actually have an environment where things are set up to work properly and the same backbone code as you were provided in answer there to ’ TQDM progress bar, change text format’ will work to at least render the progress bar as a nice widget in JupyterLab. (Note that essentially the same post was made here without linking.)
You got a notification you need ipywidgets installed and working. If you have such an environment than this code works to render an widgets-based progress bar in JupyterLab with ipywidgets installed and working with ipykernel:
from tqdm.notebook import tqdm
import time
for i in tqdm(range(5), desc='𝐁𝐨𝐥𝐝 𝐃𝐞𝐬𝐜𝐫'):
time.sleep(.4)
Details & How To demonstrate in JupyterLab (with ipykernel):
Go here and press the top ‘launch binder’ badge. Or click here to launch directly.
(Or you can use this more official offering of current JupyterLab via MyBinder from Jupyter developer Jeremy Toulup; it comes, apparently by default because curiously it isn’t listed in the requirements.txt configuration file, with ipywidgets v 8.1.5 at this time.)
When the remote-served Jupyter session comes up, open a new notebook and run the following in a cell:
%pip install tqdm
Run that cell and then restart the kernel.
Now with the kernel restarted, run the code block from your post but increase the sleep time slightly. You’ll see it use ipywidgets to render the progress bar.
(Note: it is easier to test anything involving ipywidgets where you know the environment is properly set up before launching Jupyter. Where I’ve sent you to try it, I already know & verified in both cases it gets installed and works when the session comes up.)
However, you are back to the title not being rendered bold because the implementation is different.
One work around is described here and here is it implemented:
from tqdm.notebook import tqdm
import time
for i in tqdm(range(5), desc='𝐁𝐨𝐥𝐝 𝐃𝐞𝐬𝐜𝐫 (not bold)'):
time.sleep(.4)
As for using JupyterLab from the ‘Try Jupyter’ Page, i.e., JupyterLite:
(You mean you want to know what version it is using? Or something else? Assuming something else is meant…)
Unless you need to work in JupyterLite, you are making things harder for yourself trying in JupyterLite. On the ‘Try Jupyter’ page it is clearly emphasized with prominent warning signs that JupyterLite is experimental. And if you further follow the link there to the JupyterLite page the emphasized ‘Status’ page warns yet again with prominent DANGER symbols. (In fact, pertinent to the approaches here to make a mock loop, up until recently the Pyodide kernel was known for not supporting from time import sleep, see the bottom note here in the old coverage for Voici of ‘How does xeus-python lite compare to the Pyodide kernel?’.)
With that important cautionary note out of the way…
However, at present both of the code blocks you’ll find below demonstrating tqdm with an widget bar work with JupyterLite JupyterLab where ipywidgets is first installed.
You will get an error about IProgress if ipywidgets isn’t first installed and you try to run the code though.
And in fact clearing the kernel doesn’t help.
(At this time for the Pyodide kernel you dynamically install in the namespace of the running notebook, i.e., installing once in a separate notebook won’t affect a new notebook you open in JupyterLite with the Pyodide kernel. In other words each new notebook won’t have the ipywidgets installed, unlike how it works in a proper ipykernel.)
Demonstration if using with JupyterLite JupyterLab:
Go to the ‘Try Jupyter’ page. (I suggest doing that last step in an incognito mode or in a private browser window so you get fresh JupyterLite for testing.) When that page comes up, double-click on the notebook listed entitled ‘Lorenz.ipynb’ to open it. When it comes up, let the kernel start up and then go to not busy. Now run the first cell to install ipywidgets.
Alternatively, start a new notebook with the ‘Python (Pyodide)’ as kernel choice, wait until the kernel status indicator is clear, and then run in the first cell the following to install ipywidgets in that notebook namespace:
%pip install -q ipywidgets
When the kernel status indicator again shows clear after installing with that command in either situation, run one of the following in a new cell:
from tqdm.notebook import tqdm
import asyncio
for i in tqdm(range(5), desc='𝐁𝐨𝐥𝐝 𝐃𝐞𝐬𝐜𝐫 (not bold)'):
await asyncio.sleep(0.4)
-and-
from tqdm.notebook import tqdm
tqdm.monitor_interval = 0
import time
for i in tqdm(range(5), desc='𝐁𝐨𝐥𝐝 𝐃𝐞𝐬𝐜𝐫 (not bold)'):
time.sleep(.4)
(Note if you remove the tqdm.monitor_interval = 0 line, at this time you get the warning: TqdmMonitorWarning: tqdm:disabling monitor support (monitor_interval = 0) due to: can't start new thread.)
And while it is better to do the install at first and then try, I found that at this time you can get away with running the following in a new cell in a new notebook in JupyterLite to do both steps at once:
%pip install -q ipywidgets
from tqdm.notebook import tqdm
tqdm.monitor_interval = 0
import asyncio
for i in tqdm(range(5), desc='𝐁𝐨𝐥𝐝 𝐃𝐞𝐬𝐜𝐫 (not bold)'):
await asyncio.sleep(0.4)
