Failed to start basic examples

[ad-mos]

After successfuly cloning, setting up packages and installing requirements I cant run examples scripts. Tried to run snippet from Readme.md and examples/py/rest_examples/simple_flask/server.py, both gives the same error after Adding AutoShape... step:

/usr/lib/python3/dist-packages/pkg_resources/__init__.py:116: PkgResourcesDeprecationWarning: 1.1build1 is an invalid version and will not be supported in a future release
warnings.warn(
YOLOv5 🚀 v7.0-56-gc0ca1d2 Python-3.10.6 torch-1.13.1+cu117 CPU

Fusing layers...
Model summary: 213 layers, 7012822 parameters, 0 gradients
Adding AutoShape...
Traceback (most recent call last):
File "/home/dev/nomeroff-net/test.py", line 4, in <module>
number_plate_detection_and_reading = pipeline("number_plate_detection_and_reading",
File "/home/dev/nomeroff-net/nomeroff_net/pipelines/__init__.py", line 115, in pipeline
return pipeline_class(task, image_loader, **pipeline_kwargs, **kwargs)
File "/home/dev/nomeroff-net/nomeroff_net/pipelines/number_plate_detection_and_reading.py", line 49, in __init__
self.number_plate_text_reading = NumberPlateTextReading(
File "/home/dev/nomeroff-net/nomeroff_net/pipelines/number_plate_text_reading.py", line 69, in __init__
self.detector = class_detector(presets, default_label, default_lines_count)
File "/home/dev/nomeroff-net/nomeroff_net/pipes/number_plate_text_readers/text_detector.py", line 45, in __init__
self.load()
File "/home/dev/nomeroff-net/nomeroff_net/pipes/number_plate_text_readers/text_detector.py", line 60, in load
detector.load(self.prisets[detector_name]['model_path'])
File "/home/dev/nomeroff-net/nomeroff_net/pipes/number_plate_text_readers/base/ocr.py", line 311, in load
return self.load_model(path_to_model, nn_class=nn_class)
File "/home/dev/nomeroff-net/nomeroff_net/pipes/number_plate_text_readers/base/ocr.py", line 281, in load_model
self.model = nn_class.load_from_checkpoint(path_to_model,
File "/home/dev/.local/lib/python3.10/site-packages/pytorch_lightning/core/saving.py", line 137, in load_from_checkpoint
return _load_from_checkpoint(
File "/home/dev/.local/lib/python3.10/site-packages/pytorch_lightning/core/saving.py", line 180, in _load_from_checkpoint
return _load_state(cls, checkpoint, strict=strict, **kwargs)
File "/home/dev/.local/lib/python3.10/site-packages/pytorch_lightning/core/saving.py", line 238, in _load_state
keys = obj.load_state_dict(checkpoint["state_dict"], strict=strict)
File "/home/dev/.local/lib/python3.10/site-packages/torch/nn/modules/module.py", line 1671, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for NPOcrNet:
size mismatch for linear1.weight: copying a param with shape torch.Size([512, 1952]) from checkpoint, the shape in current model is torch.Size([416, 1952]).
size mismatch for linear1.bias: copying a param with shape torch.Size([512]) from checkpoint, the shape in current model is torch.Size([416]).
size mismatch for recurrent_layer1.rnn.weight_ih_l0: copying a param with shape torch.Size([128, 512]) from checkpoint, the shape in current model is torch.Size([128, 416]).
size mismatch for recurrent_layer1.rnn.weight_ih_l0_reverse: copying a param with shape torch.Size([128, 512]) from checkpoint, the shape in current model is torch.Size([128, 416]).

Content of /home/dev/nomeroff-net/test.py:

from nomeroff_net import pipeline
from nomeroff_net.tools import unzip

number_plate_detection_and_reading = pipeline("number_plate_detection_and_reading", 
                                              image_loader="opencv")

(images, images_bboxs, 
 images_points, images_zones, region_ids, 
 region_names, count_lines, 
 confidences, texts) = unzip(number_plate_detection_and_reading(['./11.jpg', 
                                                                 './5.jpg']))
 
print(texts)

Does anybody know what am I doing wrong?

[ad-mos]

@ApelSYN maybe wrong models was autoloaded?

UPD: Need to run the code from branch v3.2.

[ApelSYN]

We prematurely committed to the branch v3.3 master branch, there are now a few pieces of code that need to be fixed, so I recommend that you run the code from branch v3.2.
I think that within a week we will correct all the errors.

[ad-mos]

It works now, many thanks!

[dimabendera]

You can try now with the new version (master branch) to remove metadata of models (folder ./nomeroff-net/data/configs) and update the branch (git pull). That should help.