Recent content by Msss

Thanks!

Have some questions on the final scores. On the codalab page, the score function would be like this. However, it don't seem like the final scores are coming from the function. e.g. 2^( 2 * 2.6 ) /28.1 = 1.308 (not 74) 2^( 2 * 3.5 ) /89.3 = 1.433 (not 28)

Has the draft of the challenge report been released?

Submission ID : 162072 Username : MinsuKwon It looks like my tflite model for track 2 was reported in the result table. Please find the email for more details about the issue!

1. There is no instruction regarding the fixed input size even though the final phase has been started. And no instruction for tflite as well. 2. It looks like it is not necessary to submit result images. Right? Thank you

How is the submission supposed to be when I am participating in both track1 and track2.

@Andrey Ignatov Hi, we are Enerzai research team and has a question. One of our solution used channel-wise attention with global average pooling and found that it induces a huge increase in latency. And that's why our one solution has very large latency. (We checked that removing the...

Okay got it! Thanks a lot

I just found the runtime sheet https://docs.google.com/spreadsheets/d/e/2PACX-1vQ-pjVpm0RXORsxP1JIAfX-IW9Hyn4jYCe-n515L08fEmAbbZ2fWntgZS5uigJos9gawt2-0bq1pLRa/pubhtml has been updated. Submission ID 833803 got runtime error even the model ran on my galaxy S21 (Exynos 2100) with all settings...

I guess the runtime measurement is extremely inaccurate. I got the 4x slower result even when I submitted the same model.

Thank you very much!! It seems working.

Thank you for response! https://github.com/aiff22/MAI-2021-Workshop/blob/main/tensorflow_to_tflite.py Here, with the official tflite converter code, I've just replaced the place holder like you said. x_ = tf.compat.v1.placeholder(tf.float32, [1, None, None, 3], name="input") However, the...

Dear organizer, It is hard to understand how the tflite model for [1, None, None, 3] input could be made since only the tflite models with "static size input" work on mobile GPUs. Also, the resize_bilinear function requires the shape of a resized tensor. What about submitting the tflite model...

Hi! I thought test dataset would be the size of 480 x 720 as you mentioned above. But the test dataset shape is the same as validation dataset (the first dataset we used). Could you please tell us what happened?

Will the test inputs be released? If it is released, some could make denoised images with a very large network and use them for training small networks.