hzfu / agnet Goto Github PK

Under which licence do you publish your implementation?

Hi,

I am trying to understand your code after reading the paper. There are several issues I am trying to deal with. First, running the code directly from main.py, I got an error stating "NameError: name 'AG_Net' is not defined". Also, running the code directly from models.py gave another error stating that "ModuleNotFoundError: No module named 'guided_filter_pytorch.guided_filter_attention'". Thus, I am not sure which one is the base file. Also, I tried to search google for more guidance but I didn't find any.

Could you please let me know what to fix to make it work?

@HzFu hi thanks for open sourcing the work , i had few queries

1. can we use this source code to train on other datasets like autonomus driving dataset
2. can we integrated AG net with other architecture to improve your the accuracy

Please share your thoughts on this

Hi, where can I download ORIGA DataSet?

I am not sure the specific tf function that leads to this, please how can I retrace the warning to catch it?

C:\Users\Omisore\AppData\Roaming\Python\Python37\Lib\venv\scripts\python.exe "G:/Documents/Asst Prof/Publications/ICRA 2022/Code/selfTrial.py"
2021-09-11 12:41:28.922000: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN)to use the following CPU instructions in performance-critical operations: AVX AVX2
To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
Exploring [-0.08397679 -0.46502669 -0.97932531]
0
Completed in 0 trials
I am here
WARNING:tensorflow:From E:\ProgramData\Anaconda3\envs\tf23\lib\site-packages\tensorflow\python\training\tracking\tracking.py:111: Model.state_updates (from tensorflow.python.keras.engine.training) is deprecated and will be removed in a future version.
Instructions for updating:
This property should not be used in TensorFlow 2.0, as updates are applied automatically.
2021-09-11 12:41:29.289000: W tensorflow/python/util/util.cc:348] Sets are not currently considered sequences, but this may change in the future, so consider avoiding using them.
WARNING:tensorflow:From E:\ProgramData\Anaconda3\envs\tf23\lib\site-packages\tensorflow\python\training\tracking\tracking.py:111: Layer.updates (from tensorflow.python.keras.engine.base_layer) is deprecated and will be removed in a future version.
Instructions for updating:
This property should not be used in TensorFlow 2.0, as updates are applied automatically.
I got here
I got here 1
I got here 2
I got here 3

Process finished with exit code 0

Dear HzFu:
Thanks for your general work for Retinal Vessel Segmentation. In reading your code, I don't understand the way to generate train.pkl and test.pkl file, could you please help, thanks a lot!!

I wanna compare your method, Can you give me the result of image which test on DRIVE and CHASE

Hello, where is the image pre-process part and gamma correction code, I read them in your article but not found yet.

Have you divided the verification set on the DRIVE dataset? Because there are only train_list and test_list in the train_dict.pkl file, but I see an option for vali_list in the code.

Dear HzFu:
Have you ever met with the following ERROR?

Traceback (most recent call last):
  File "test.py", line 144, in <module>
    fast_test(model, args, test_img_list, model_name)
  File "test.py", line 102, in fast_test
    Auc = roc_auc_score(tmp_gt, tmp_out)
  File "/site-packages/sklearn/metrics/_ranking.py", line 381, in roc_auc_score
    raise ValueError("multi_class must be in ('ovo', 'ovr')")
ValueError: multi_class must be in ('ovo', 'ovr')

So puzzled????