satojkovic

@NikolayStarikov What happens when you run train_deep_logo_cnn.py with no args? ```csh $ python train_deep_logo_cnn.py ```

Hi @gtuvshintugs It might be helpful whole discussions in issue #4 but could you paste all error messages ?

Hi @jbogp I've evaluated DeepLogo using mean average precision because Tensorflow Object Detection API provides an evaluation script for this metric. Do you know any publicly available script for global...

@jbogp Never mind. Here is my evaluation result. PascalBoxes_Precision/[email protected]: 0.858025 If it's possible, could you share your evaluation results when you finished training? Thanks!

Hi @jbogp Very interesting results! It might be a good idea to train with larger dataset than flickr_logos_27 when you use bigger models.

Hi @TacoTijuana When I evaluated the trained model using eval.py, I got the loss values below. | classification loss | localization loss | |:-:|:-:| | 2.721 @ 6.046k step |...

hi @yumiaoGitHub Here it is. https://github.com/tensorflow/models/blob/master/research/object_detection/legacy/eval.py

Hi @yumiaoGitHub I added evaluation section to README. Could you please check?

@peckjon First of all, thank you for your PR. * LGTM with 87c528f30e123b54f018ba1536abbe9614c1f3a9. * I think this repository should not include flickr_27_logos dataset because the dataset is distributed by National...

@rakashi It seems that Pillow 2.3.0 misses the close method. Please upgrade Pillow.