Hamza Khalil comments

Results 13 comments of


Hamza Khalil

Class static used to pass info between stream steps

@jukkakarvanen Hey! im a newbie at this... im trying to run this repo using mvn clean package. Do I have to use kafka as well? or just use mvn clean...

Got the Top-1 accu of only 70.7% when training efficientnet-b3 with Imagenet

@YvetteLaw can you please share the dataset structure?

On new test image, no output

@varunagrawal do you have any visualization script?

On new test image, no output

@varunagrawal Thankyou I found it. Although the results aren't satisfactory with the 50 epochs model , do you have a more trained model? If yes kindly share.

On new test image, no output

@varunagrawal Okay , No issues thankyou..😀

Hello, what is the training data set? Where can I download it?

@llzhaoshuo @xiaoweidao Training Data can be downloaded from https://github.com/cvdfoundation/google-landmark and yes only images are required.

How to run captioning on any image? E.g. how to prepare the test.yaml and other required files required by the run_captioning.py?

@jontooy @eslambakr Hey! iv created the dataset using the three images present inside the above colab notebook... prepared dataset using type=Test and caption=False. model iv used for oscar+ is checkpoiint_base...

How to run captioning on any image? E.g. how to prepare the test.yaml and other required files required by the run_captioning.py?

never mind got it fixed.

Inference

@Kangzf1996 hey! do have code for it?