daipuwei/cgan-mnist

这是利用条件GAN(CGAN)实现生成MNIST图像的代码

I'm Daipuwei

• 🎓 辽宁石油化工大学
• 📦 上市公司算法工程师 & 前B轮安防创业公司算法总监
• 🔥DataStructure: https://github.com/Daipuwei/DataStructure
• 🔥 DANN-MNIST: https://github.com/Daipuwei/DANN-MNIST
• 🔥《机器学习入门：基于数学原理的Python实战》源码: https://github.com/Daipuwei/Introduction-to-Machine-Learning-Based-on-Mathematical-Principles-with-Python
• 🔥CGAN-mnist: https://github.com/Daipuwei/CGAN-mnist
• 🔥YOLO_Deploy: https://github.com/Daipuwei/YOLO_Deploy
• 🔥Object_Detection_Dataset_Utils: https://github.com/Daipuwei/Object_Detection_Dataset_Utils
• 📚 我的CSDN博客（CSDN Blog） https://daipuweiai.blog.csdn.net/
• 🍱 我的知乎（Zhihu） https://www.zhihu.com/people/daipuweiai

cgan-mnist's Issues

train.py error

Hello, excuse me. When I run train.py, I always get an error in epoch 32. What is the reason?

请问老师给的h5文件是可以直接load的吗? 我这边进行了如下报错, 不知道问题在哪里

Traceback (most recent call last):
File "d:\work_paul\11_new_workspace\a_inference\CGAN-mnist-master\ceshi.py", line 11, in
cgan = CGAN(cfg,weight_path)
File "d:\work_paul\11_new_workspace\a_inference\CGAN-mnist-master\CGAN\CGAN.py", line 57, in init
self.cgan.load_weights(weight_path,by_name=True)
File "D:\softwres\py\lib\site-packages\keras\utils\traceback_utils.py", line 70, in error_handler
raise e.with_traceback(filtered_tb) from None
File "D:\softwres\py\lib\site-packages\keras\saving\hdf5_format.py", line 923, in load_weights_from_hdf5_group_by_name
raise ValueError(
ValueError: Weight count mismatch for layer #3 (named model_1). Layer expects 21 weight(s). Received 9 saved weight(s)

奇怪的错误IndexError

每次epoch到32次都提示这个错误
IndexError Traceback (most recent call last)
in ()
13
14 if name == 'main':
---> 15 run_main()

1 frames
in train(self, train_datagen, epoch, k, batch_size)
213 if ep % self.config.save_epoch_interval == 0:
214 model_cgan = "Epoch{}dcgan_loss{}discriminator_loss{}acc{}.h5".format(ep, np.average(cgan_losses),
--> 215 np.average(d_losses,0)[0],np.average(d_losses,0)[1])
216 self.cgan.save(os.path.join(model_path, model_cgan))
217 save_dir = os.path.join(train_result_path, str("Epoch{}".format(ep)))

IndexError: invalid index to scalar variable.