EOFError: Ran out of input

[EvansXu]

The error message I receive is as follows:

evans@evans:~/Work/Rerender_A_Video$ python3 rerender.py --cfg config/real2sculpture.json
logging improved.
Traceback (most recent call last):
File "rerender.py", line 25, in
from flow.flow_utils import get_warped_and_mask
File "/home/evans/Work/Rerender_A_Video/flow/flow_utils.py", line 258, in
flow_calc = FlowCalc()
File "/home/evans/Work/Rerender_A_Video/flow/flow_utils.py", line 156, in init
checkpoint = torch.load(model_path,
File "/usr/local/lib/python3.8/dist-packages/torch/serialization.py", line 1028, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/usr/local/lib/python3.8/dist-packages/torch/serialization.py", line 1246, in _legacy_load
magic_number = pickle_module.load(f, **pickle_load_args)
EOFError: Ran out of input

evans@evans:~/Work/Rerender_A_Video$ nvidia-smi
Sun Nov 5 22:02:46 2023
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.129.03 Driver Version: 535.129.03 CUDA Version: 12.2 |
|-----------------------------------------+----------------------+----------------------+
| GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. |
| | | MIG M. |
|=========================================+======================+======================|
| 0 NVIDIA TITAN Xp Off | 00000000:65:00.0 On | N/A |
| 23% 34C P5 23W / 250W | 727MiB / 12288MiB | 1% Default |
| | | N/A |
+-----------------------------------------+----------------------+----------------------+

+---------------------------------------------------------------------------------------+
| Processes: |
| GPU GI CI PID Type Process name GPU Memory |
| ID ID Usage |
|=======================================================================================|
| 0 N/A N/A 865 G /usr/lib/xorg/Xorg 102MiB |
| 0 N/A N/A 1469 G /usr/lib/xorg/Xorg 312MiB |
| 0 N/A N/A 1596 G /usr/bin/gnome-shell 92MiB |
| 0 N/A N/A 3353 G /usr/lib/firefox/firefox 204MiB |
+---------------------------------------------------------------------------------------+

[williamyang1991]

The Python "Pickle EOFError: Ran out of input" occurs when the file you are trying to open and load is empty.
To solve this error, you need to make sure all the pretrained models are downloaded completely.

Your error indictate the following model is not correctly downloaded.

Rerender_A_Video/install.py

Lines 50 to 53 in a348e0a

	def download_gmflow_ckpt():
	url = ('https://huggingface.co/PKUWilliamYang/Rerender/'
	'resolve/main/models/gmflow_sintel-0c07dcb3.pth')
	download(url, 'models')

[EvansXu]

@williamyang1991 Thanks for your enthusiastic reply!
After using python install.py, the result shows success without any error prompts. I can see the following information under the ./model. Is it correct?
evans@evans-MS-7A94:~/Work/Rerender_A_Video/models$ ls -l
total 0
-rw-rw-r-- 1 evans evans 0 11月 5 20:09 control_sd15_canny.pth
-rw-rw-r-- 1 evans evans 0 11月 5 20:09 control_sd15_hed.pth
-rw-rw-r-- 1 evans evans 0 11月 5 20:09 gmflow_sintel-0c07dcb3.pth
-rw-rw-r-- 1 evans evans 0 11月 5 20:09 vae-ft-mse-840000-ema-pruned.ckpt

by the way, In file real2sculpture.json, it was found that file models/realisticVisionV20_v20.safetensors is needed ,but I can't find the file realisticVisionV20_v20.safetensors, May I ask if the error was caused by a missing rrealisticVisionV20_v20.safetensors. Thansk a lot!

[williamyang1991]

Yes, you can download realisticVisionV20_v20.safetensors here

[EvansXu]

@williamyang1991 Thanks, I'll download it and have a try!