Emmanuel Benazera

You really need to post all docker and API *calls*, *output logs*, etc... so someone can try to reproduce to help you. Otherwise if you put yourself in our seat,...

Sure, it is actually pretty easy. - First setup your dataset, the format is described here: https://www.deepdetect.com/platform/docs/#object-detection It's basically a text file with image path and bounding box file path....

Ah, see #765. I had tested in-memory training only. To add svm + db training to unit tests I had to add `test_split` support to SVM + db training first,...

Hi, libpng says it, there's an issue with an image somewhere. Best way is to write a script that decodes all images to decode all images. To debug if it's...

Ah hi @YaYaB good catch, as usual! So one way we have here is that we only store an id, or the relevant part of the URI, this is why...

OK, regarding the warning, we may allow it only once, because otherwise, it'd get printed millions of times for large datasets for which most of the time, the id in...

See if #755 is fine for your need.

Interesting thanks, @fantes can you double check on this ?

@YaYaB Can you try `caffe::EncodeCVMatToDatum` here in place of the `CVMatToDatum`, https://github.com/jolibrain/deepdetect/blob/master/src/backends/caffe/caffeinputconns.h#L213 ? Typically, like this: ```C++ caffe::EncodeCVMatToDatum(this->_images.at(i),guess_encoding(_uris.at(i)),&datum); ``` We use `EncodeCVMatToDatum` when creating the LMDBs to not store potentially...

Hi @YaYaB, this is because your model is overfitted, with high certainty. Actually, the difference of prediction between the encoded and raw version of images is also a consequence of...