OSError: unknown file format #149
Comments
|
Hi! Please verify that the files are accessible in the right directory and that, if using a networked source like GDrive, the connection is active. |
|
for say true we got this on this line i try this one too |
|
This line:
Makes me think you might be using an outdated version. Try this one: https://colab.research.google.com/github/AlphaGit/DAIN/blob/master/Colab_DAIN.ipynb I have included several updates and fixes in it, which should match the version from here: https://github.com/AlphaGit/DAIN Let me know if it helps! |
|
look finish on same error at first WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv about front import ` Interpolation%shell mkdir -p '{FRAME_OUTPUT_DIR}' !python -W ignore colab_interpolate.py --netName DAIN_slowmotion --time_step {fps/TARGET_FPS} --start_frame 1 --end_frame {frame_count} --frame_input_dir '{FRAME_INPUT_DIR}' --frame_output_dir '{FRAME_OUTPUT_DIR}' --mixer={FRAME_MIXER} --resize_hotfix {RESIZE_HOTFIX} so i have change the colab interpotale for import manualy big_font = ImageFont.truetype("/content/DAIN/Roboto-Bold.ttf"), size=24)
|
|
Hi! Sorry, I don't fully understand how this part is related:
Are you manually importing fonts? What are you doing with them? |
|
yes manually importing fonts for try fix the File "colab_interpolate.py", line 64 |
|
I see -- those fonts are only used to write over the image and debug intermediate interpolation results. Feel free to comment out:
And replaces the contents of the the d debug method (lines 339 to 342) with the word "pass". This would take out all that interaction and still allow the rest of the notebook to run. |
|
got one frame result and assertionError /content/DAIN have try and retry |
|
Good, progress!
It seems that your image files have 4 channels instead of 3. DAIN can only
work with 3. Maybe they have a transparency layer?
Make sure that the part of the code that removes the alpha layer is
running.
…On Wed, Jan 26, 2022 at 4:41 AM kilik128 ***@***.***> wrote:
got one frame result and assertionError
/content/DAIN
revise the unique id to a random numer 65014
Namespace(SAVED_MODEL=None, alpha=[0.0, 1.0],
arg='./model_weights/65014-Wed-Jan-26-09-36/args.txt', batch_size=1,
channels=3, ctx_lr_coe=1.0, datasetName='Vimeo_90K_interp', datasetPath='',
dataset_split=97, debug=False, depth_lr_coe=0.001,
doubleframe_distorted_pixel_ratio=0.002,
doubleframe_distortion_threshold=20, dtype=<class
'torch.cuda.FloatTensor'>, end_frame=36, epsilon=1e-06, factor=0.2,
filter_lr_coe=1.0, filter_size=4, flow_lr_coe=0.01, force=False,
frame_input_dir='/content/DAIN/input_frames',
frame_output_dir='/content/DAIN/output_frames',
log='./model_weights/65014-Wed-Jan-26-09-36/log.txt', lr=0.002,
mixer='normal', netName='DAIN_slowmotion', no_date=False, numEpoch=100,
occ_lr_coe=1.0, patience=5, rectify_lr=0.001, resize_hotfix=True,
save_path='./model_weights/65014-Wed-Jan-26-09-36', save_which=1, seed=1,
start_frame=1, time_step=0.1998001998001998, uid=None, use_cuda=True,
use_cudnn=1, weight_decay=0, workers=8)
cudnn is used
Interpolate 4 frames
3% 1/35 [00:00<00:00, 41.87it/s]
Traceback (most recent call last):
File "colab_interpolate.py", line 397, in
interpolate_mixer(input_frame)
File "colab_interpolate.py", line 213, in normal_interpolate
interpolate(input_frame, input_frame+1)
File "colab_interpolate.py", line 205, in interpolate
for frame, time_offset in zip(subframes(image_1, image_2), time_offsets):
File "colab_interpolate.py", line 125, in subframes
assert channels == 3
AssertionError
have try and retry
—
Reply to this email directly, view it on GitHub
<#149 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAIP2WMFTOCMTPEU34C74GTUX66TVANCNFSM5MTJB6CA>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
|
Done Thank's ! it's realy cool :) |
got this
/content/DAIN
revise the unique id to a random numer 67519
Namespace(SAVED_MODEL=None, alpha=[0.0, 1.0], arg='./model_weights/67519-Sun-Jan-23-12-26/args.txt', batch_size=1, channels=3, ctx_lr_coe=1.0, datasetName='Vimeo_90K_interp', datasetPath='', dataset_split=97, debug=False, depth_lr_coe=0.001, doubleframe_distorted_pixel_ratio=0.002, doubleframe_distortion_threshold=20, dtype=<class 'torch.cuda.FloatTensor'>, end_frame=36, epsilon=1e-06, factor=0.2, filter_lr_coe=1.0, filter_size=4, flow_lr_coe=0.01, force=False, frame_input_dir='/content/DAIN/input_frames', frame_output_dir='/content/DAIN/output_frames', log='./model_weights/67519-Sun-Jan-23-12-26/log.txt', lr=0.002, mixer='normal', netName='DAIN_slowmotion', no_date=False, numEpoch=100, occ_lr_coe=1.0, patience=5, rectify_lr=0.001, resize_hotfix=False, save_path='./model_weights/67519-Sun-Jan-23-12-26', save_which=1, seed=1, start_frame=1, time_step=0.11988011988011989, uid=None, use_cuda=True, use_cudnn=1, weight_decay=0, workers=8)
cudnn is used
Interpolate 7 frames
Traceback (most recent call last):
File "colab_interpolate.py", line 64, in
big_font = ImageFont.truetype(urlopen(font_url), size=24)
File "/usr/local/lib/python3.7/dist-packages/PIL/ImageFont.py", line 648, in truetype
return freetype(font)
File "/usr/local/lib/python3.7/dist-packages/PIL/ImageFont.py", line 645, in freetype
return FreeTypeFont(font, size, index, encoding, layout_engine)
File "/usr/local/lib/python3.7/dist-packages/PIL/ImageFont.py", line 197, in init
load_from_bytes(font)
File "/usr/local/lib/python3.7/dist-packages/PIL/ImageFont.py", line 179, in load_from_bytes
"", size, index, encoding, self.font_bytes, layout_engine
OSError: unknown file format
on !python -W ignore colab_interpolate.py --netName DAIN_slowmotion --time_step {fps/TARGET_FPS} --start_frame 1 --end_frame {pngs_generated_count} --frame_input_dir '{FRAME_INPUT_DIR}' --frame_output_dir '{FRAME_OUTPUT_DIR}'
The text was updated successfully, but these errors were encountered: