Galuh Putra Warman
Galuh Putra Warman
is there a different dataset from between dataset in this repo and the http://shuoyang1213.me/WIDERFACE/index.html??
in colab ``` %cd /content/MTCNN-Tensorflow/data/MTCNN_model %mkdir PNet_No_Landmark %mkdir ONet_No_Landmark %mkdir RNet_No_Landmark import shutil #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_landmark/ONet-16.data-00000-of-00001", "/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_No_landmark/ONet-16.data-00000-of-00001") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_landmark/ONet-16.index", "/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_No_landmark/ONet-16.index") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_landmark/ONet-16.meta", "/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_No_landmark/ONet-16.meta") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_landmark/checkpoint", "/content/MTCNN-Tensorflow/data/MTCNN_model/ONet_No_landmark/checkpoint") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_landmark/PNet-18.data-00000-of-00001", "/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_No_landmark/PNet-18.data-00000-of-00001") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_landmark/PNet-18.index", "/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_No_landmark/PNet-18.index") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_landmark/PNet-18.meta", "/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_No_landmark/PNet-18.meta") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_landmark/checkpoint", "/content/MTCNN-Tensorflow/data/MTCNN_model/PNet_No_landmark/checkpoint") #shutil.copy("/content/MTCNN-Tensorflow/data/MTCNN_model/RNet_landmark/RNet-14.data-00000-of-00001",...
sam with me
i think in gen_hard_example.py we are using Onet but we must use RNet first so change it in line around 231-240 `if __name__ == __main__: #net = 'ONet' net =...
oh man i cant really understand u guys i have this problem , can u guys explain to me briefly how to solve this issue? thanks
im already in this step for pnet? Step 808673 for cls: loss = 0.15127. Step 808526 for bbx: loss = 0.00650. Step 1 for pts: loss = 0.00000
PNet: Step 1547385 for cls: loss = 0.13956. Step 1548614 for bbx: loss = 0.00630. Step 1 for pts: loss = 0.00000 Done training for 200 epochs, 1547408 steps. about...
change line 77 into this: try: neg_keep = npr.choice(len(neg), size=1000000, replace=False) except ValueError: neg_keep = npr.choice(len(neg), size=1000000, replace=True)
im solving this by using the !wget inception model and !cp line of code from this website: https://www.kaggle.com/code/nulldata/pixel-like-unblur-image-deblur-with-deblurganv2/notebook