search

tianyili2017 / HEVC-Complexity-Reduction

Source programs to test the deep-learning-based complexity reduction approach for HEVC, at both intra- and inter-modes.
124 stars 80 forks source link

训练模型问题 #22

Open qin-liming opened 3 years ago

qin-liming commented 3 years ago

请问一下从小样本换成全部样本训练时，需要注意input_data.py和train_CNN_CTU64.py中哪里的修改，只修改了训练样本数量可以吗，训练报错在images=data [:, 0:4096].astype(np.float32),MemoryError: Unabale to allocate 1.22Gib for an array with shape (80000,4096)and data type float32

tianyili2017 commented 3 years ago

您好，已经在另外一个issue上发现同样问题，并回复。谢谢您关注我们的工作！

CherryXChen commented 3 years ago

您好，想问一下您用全部数据训练的时候会不会出现loss变成nan的情况