Tqdm: Cannot redraw tqdm_notebook Hbox progress bar after calling clear_output(), only ascii shown

[phillies]

• [x] I have visited the [source website], and in particular
  read the [known issues]
• [x] I have searched through the [issue tracker] for duplicates
• [x] I have mentioned version numbers, operating system and
  environment, where applicable:
  python import tqdm, sys print(tqdm.__version__, sys.version, sys.platform) #4.36.1 3.7.4 (default, Aug 13 2019, 20:35:49) [GCC 7.3.0] linux

When I clean the output of a cell during runtime with IPython.display.clear_output() and want to re-draw the tddm_notebook progress bar I only get an ascii representation instead of the HBox. I'm not sure if this is a bug or not possible by design. A simple example here:

%matplotlib inline
import time
import numpy as np
import matplotlib.pyplot as plt
from tqdm import tqdm_notebook

from IPython.display import display, clear_output, update_display

fig, ax = plt.subplots(1, 1)
style = ['bo', 'rs', 'gd']

progress_bar = tqdm_notebook(iterable=range(50), leave=True)

for ii in progress_bar:
    for jj in range(3):
        ax.plot(ii, np.random.randn(1), style[jj])

    clear_output(wait=True)
    display(fig)
    display(progress_bar)
    plt.pause(1)

This should display a plot and a progress bar which are both updated on every iteration. The figure is shown correctly, the progress bar just as ascii. Calling .display() or .refresh() on the tqdm_notebook progress bar has no effect.

[casperdcl]

think this is a jupyter thing. You get the same effect using e.g. matplotlib figures instead of tqdm progress bars.

The object still exists but you've cleared its display so can't see it.

[casperdcl]

happy to re-open if there's anything we can do.

[nokados]

According to this issue https://github.com/jupyter-widgets/ipywidgets/issues/1744
you can only erase your plot:

import time
import numpy as np
import matplotlib.pyplot as plt
from tqdm.auto import trange

from IPython import display
from ipywidgets import Output

out = Output()
display.display(out)
for i in trange(10):
    with out:
        plt.plot(np.random.randn(100))
        plt.show()
        display.clear_output(wait=True)
    time.sleep(1)