Better explanation or improvement of default tqdm bar?

[brando90]

I was seeing:

n_epoch=2: train_loss=0.8231842517852783 train_acc=0.02586206896551724
n_epoch=2: val_loss=0.7130754590034485 val_acc=0.5546875
  3%|█████████▌                                                                                                                                                                                                                                                                                                                      | 3/100 [00:44<15:26,  9.55s/it]
n_epoch=3: train_loss=0.796868085861206 train_acc=0.4396551724137931
n_epoch=3: val_loss=0.6613789200782776 val_acc=0.5546875
  6%|███████████████████▏                                                                                                                                                                                                                                                                                                            | 6/100 [01:00<10:42,  6.84s/it]
n_epoch=4: train_loss=0.7267095446586609 train_acc=0.017241379310344827
n_epoch=4: val_loss=0.6583153605461121 val_acc=0.0
 10%|███████████████████████████████▉                                                                                                                                                                                                                                                                                               | 10/100 [01:15<07:50,  5.23s/it]
n_epoch=5: train_loss=0.7636414170265198 train_acc=0.4396551724137931
n_epoch=5: val_loss=0.6460121273994446 val_acc=0.5546875
 15%|███████████████████████████████████████████████▊                                                                                                                                                                                                                                                                               | 15/100 [01:30<05:54,  4.17s/it]
n_epoch=6: train_loss=0.7080773711204529 train_acc=0.05172413793103448
n_epoch=6: val_loss=0.6564346551895142 val_acc=0.0
 21%|██████████████████████████████████████████████████████████████████▉                                                                                                                                                                                                                                                            | 21/100 [01:45<04:27,  3.39s/it]

which was confusing because I cant find what the notation [01:15<07:50 is supposed to mean.

Also, the text is put really far away at the end of my screen on my ultra wide monitor...isn't it possible to display it more like this:

n_epoch=91: train_loss=0.3727933168411255 train_acc=0.6810344827586207
n_epoch=91: val_loss=0.29822027683258057 val_acc=0.7421875
 91% (91 of 100) |####################   | Elapsed Time: 0:22:40 ETA:   0:02:12

[casperdcl]

tqdm(
    ...,
    bar_format="{percentage:3.0f}% ({n_fmt} of {total_fmt}) |{bar:23a}| Elapsed Time: {elapsed:>7} ETA: {remaining:>7}")

[brando90]

tqdm(
    ...,
    bar_format="{percentage:3.0f}% ({n_fmt} of {total_fmt}) |{bar:23a}| Elapsed Time: {elapsed:>7} ETA: {remaining:>7}")

what about this bar (implemented in progressbar 2 for concreteness):

import time
import progressbar
widgets = [
    progressbar.Percentage(),
    ' ', progressbar.SimpleProgress(format=f'({progressbar.SimpleProgress.DEFAULT_FORMAT})'),
    ' ', progressbar.Bar(),
    ' ', progressbar.Timer(), ' | '
                              ' ', progressbar.ETA(), ' | ',
    ' ', progressbar.AdaptiveTransferSpeed(unit='it'),
]
bar = progressbar.ProgressBar(widgets=widgets)
for i in bar(range(100)):
    time.sleep(0.1)
    bar.update(i)
100% (100 of 100) |#####| Elapsed Time: 0:00:10 |  Time:  0:00:10 |    9.8 it/s

ref: wolph/python-progressbar#253

[brando90]

tqdm(
    ...,
    bar_format="{percentage:3.0f}% ({n_fmt} of {total_fmt}) |{bar:23a}| Elapsed Time: {elapsed:>7} ETA: {remaining:>7}")

also an issue I've had in the past is that the tqdm bar tries to use my entire ultra wide screen and then I can't actually see the progress bar well. Is it possible to "indent it to the left" or something like that?

[casperdcl]

super easy to do this. You may want to consult the standard Python format string syntax

[npcddyksksks]

草泥马