-
# Summary
BF16 Softmax on EMR got 40% performance regression after V3.5 "enable scratchpad for non-f32 dst dt"
Below is the reproducing step, we can see that after commit ad99c73b3ee195e4d60384a6bf5…
-
Hello @rasbt,
first of all thanks for making all this material available online, as well as your video lectures! A really helpful resource!
A small issue and fix: The classic softmax regression…
-
-
Thanks for sharing your code!!
I noticed that in your paper you mentioned that your softmin (i.e., softmax(-cost)) whereas in your code you use softmax (i.e., softmax(cost)). Am I missing anything …
-
-
I don't see any logistic function code in 2_logistic_regression.py, is this really code for logistic regression?
-
Hello,
Thank you very much for sharing your interesting work. I tried to reproduce the numbers on the pouring dataset using the following config on a machine with 4 GPUs with batch size of 18 per G…
-
pytorch-tutorial/tutorials/01-basics/logistic_regression/main.py
`
model = nn.Linear(input_size, num_classes)
`
`
criterion = nn.CrossEntropyLoss()
`
-
**bug描述**
3.6_softmax-regression-scratch.ipynb "3.6.8 预测"下的cell运行错误:
RuntimeError Traceback (most recent call last)
in
2
3 true_labels = d2l.get_fashio…
-
我的 softmax 在计算 loss 和梯度的时候都发生了数据溢出的情况，请问有什么解决方法吗（我尝试调小 learning rate，调小 reg，还有调小 num_iters 但是这些都会使 softmax 的效果下降，我尝试改用更高精度的数据类型也依然发生溢出😢😭）