always download model itself.

I am new to EasyOCR. As the example in Readme.MD, using pip install easyocr, and runing the example. I download the model from model hub, and put into ~/.EasyOCR/model,

$ ls -1 ~/.EasyOCR/model/
english_g2.zip
temp.zip
zh_sim_g2.zip

but, each time I run the code, it always show the downloading. My network is not good, so, the download always interrupted.

$ python3 test1.py
CUDA not available - defaulting to CPU. Note: This module is much faster with a GPU.
Downloading detection model, please wait. This may take several minutes depending upon your network connection.
Progress: |████----------------------------------------------| 10.0% Complet
Traceback (most recent call last):
  File "test1.py", line 5, in <module>
    reader = easyocr.Reader(['ch_sim','en']) # this needs to run only once to load the model into memory
  File "/usr/local/lib/python3.6/site-packages/easyocr/easyocr.py", line 90, in __init__
    download_and_unzip(detection_models[detector_model]['url'], detection_models[detector_model]['filename'], self.model_storage_directory, verbose)
  File "/usr/local/lib/python3.6/site-packages/easyocr/utils.py", line 586, in download_and_unzip
    urlretrieve(url, zip_path, reporthook=reporthook)
  File "/usr/lib64/python3.6/urllib/request.py", line 289, in urlretrieve
    % (read, size), result)
urllib.error.ContentTooShortError: <urlopen error retrieval incomplete: got only 9188135 out of 77251756 bytes>

Is there any wrong ? Do I need unzip the file ?

[q1208c]

Unzip both the text detection models and the language models and keep them within the same directory. Mention the path of this directory while creating a easyocr.Reader object by specifying the model_storage_directory parameter. Finally, set download_enabled to False. The last step is the most important because Easyocr will always prefer to download models from its online repository instead of searching the local file system.

Thank for your help. It’s working now.

[pronoym99]

Unzip both the text detection models and the language models and keep them within the same directory. Mention the path of this directory while creating a easyocr.Reader object by specifying the model_storage_directory parameter. Finally, set download_enabled to False. The last step is the most important because Easyocr will always prefer to download models from its online repository instead of searching the local file system.