size mismatch for Prediction.bias Error!
Chiyan200 opened this issue · 2 comments
This code is used
import easyocr
reader = easyocr.Reader(['ta'], gpu=False) # this needs to run only once to load the model into memory
result = reader.readtext('tamTxt.png')
print("result : ",result)
i got this error i run this code using cpu
Traceback (most recent call last):
File "C:\python\test\g.py", line 2, in
reader = easyocr.Reader(['ta'], gpu=False) # this needs to run only once to load the model into memory
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Users\Python312\Lib\site-packages\easyocr\easyocr.py", line 231, in init
self.recognizer, self.converter = get_recognizer(recog_network, network_params,
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "C:\Python\Python312\Lib\site-packages\easyocr\recognition.py", line 174, in get_recognizer
model.load_state_dict(new_state_dict)
File "C:\Python\Python312\Lib\site-packages\torch\nn\modules\module.py", line 2215, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for Model:
size mismatch for Prediction.weight: copying a param with shape torch.Size([143, 512]) from checkpoint, the shape in current model is torch.Size([127, 512]).
size mismatch for Prediction.bias: copying a param with shape torch.Size([143]) from checkpoint, the shape in current model is torch.Size([127]).
in https://github.com/JaidedAI/EasyOCR/releases/tag/v1.1.7
Tamil character is number + symbol + en_char + ta_char
Lines 194 to 200 in 914b164
Lines 49 to 51 in 914b164
EasyOCR/easyocr/character/ta_char.txt
Lines 1 to 47 in 914b164
The values for characters are as follows, with a total of 144.
0123456789!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~ abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZஃஅஆஇஈஉஊஎஏஐஒஓஔகஙசஜஞடணதநனபமயரறலளழவஷஸஹாிீுூெேைொோௌ்
and in https://github.com/JaidedAI/EasyOCR/releases/tag/v1.7.2
Lines 127 to 134 in c4f3cd7
The values for characters are as follows, with a total of 128.
0123456789!"#$%&\'()*+,-./:;<=>?@[\\]^_`{|}~ abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZஃஅஆஇஈஉஊஎஏஐஒஓஔகஙசஜஞடணதநனபமயரறலளழ
This code is used import easyocr reader = easyocr.Reader(['ta'], gpu=False) # this needs to run only once to load the model into memory result = reader.readtext('tamTxt.png') print("result : ",result)
i got this error i run this code using cpu Traceback (most recent call last): File "C:\python\test\g.py", line 2, in reader = easyocr.Reader(['ta'], gpu=False) # this needs to run only once to load the model into memory ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\Python312\Lib\site-packages\easyocr\easyocr.py", line 231, in init self.recognizer, self.converter = get_recognizer(recog_network, network_params, ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Python\Python312\Lib\site-packages\easyocr\recognition.py", line 174, in get_recognizer model.load_state_dict(new_state_dict) File "C:\Python\Python312\Lib\site-packages\torch\nn\modules\module.py", line 2215, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( RuntimeError: Error(s) in loading state_dict for Model: size mismatch for Prediction.weight: copying a param with shape torch.Size([143, 512]) from checkpoint, the shape in current model is torch.Size([127, 512]). size mismatch for Prediction.bias: copying a param with shape torch.Size([143]) from checkpoint, the shape in current model is torch.Size([127]).
i also getting this error so please tell me how to solve this