Fix onnx prediction

[MateoLostanlen]

Hi, I hadn't done a test with images of different sizes and I realized that the resize was not adapted. We don't need a letterbox (as is the case when training) but simply the following resizing.

I modified the model in onnx format on hugging face by exporting with the “dynamic” parameter to adapt to any image size. The dimensions simply have to be a multiple of 32

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 87.84%. Comparing base (45b19e0) to head (75d7466).

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #202      +/-   ##
===========================================
- Coverage    87.92%   87.84%   -0.09%     
===========================================
  Files            6        6              
  Lines          381      362      -19     
===========================================
- Hits           335      318      -17     
+ Misses          46       44       -2     

Flag	Coverage Δ
unittests	87.84% <100.00%> (-0.09%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.