-
in AlexNet ,
the order of the first and second layers is ( conv——LRN——maxPool )
but in your code
`network = input_data(shape=[None, 227, 227, 3])`
`network = conv_2d(network, 96, 11, strides=4…
-
I am trying to implement FCN from scratch on Pascal VOC 2012 after studying the [paper.](https://arxiv.org/pdf/1605.06211.pdf) But every time I try to train it memory goes out. I tried :
Reducing t…
-
root@1a89b5aa5fce:/opt/hipBLASLt/build/release# ./clients/staging/hipblaslt-bench -m 2048 -n 2048 -k 2048 --precision f32_r -v 1 --activation_type relu
Query device success: there are 1 devices
----…
-
I have downloaded weights checkpoints as the tutorial says, and I run vgg8 / alexnet very well, but when I run lenet or vgg16, it corrupts with an error saying,
```
File "main.py", line 156, in
…
-
Hi,
Was the model trained on relu+scale combination or it was trained on prelu and you just replace the prelu with equivalent operations only in the prototxt file.
I have a model which was trained …
-
Hi, I was about to solve a multi-dimensional Stochastic PDE like the one that used to be in the documentation. No wonder why you removed it, the old code does not work. Therefore, what modifications d…
-
I am trying to use two GPUs for training cifar10 dataset. But I got this error `File "cifar.py", line 58, in
with tf.arg_scope([tflearn.variables.variable], device='/cpu:0'):
AttributeError: mo…
-
请问转换模型过程中出现bmcpu init: skip cpu_user_defined
Cannot open libusercpu.so, disable user cpu layer.最后虽然生成了bmodel模型。但是好像relu和backword跳过了
xc597 updated
3 months ago
-
While writing the inference code for NetworkA, I encountered an issue with the Flatten layer not calculating the shape correctly during the model construction. Here is my code:```python
import tensor…
-
I would like to pass the output of a VGG16 feature extractor (-1, 512, 7, 7) onto 2 sequential-identical convLSTM layers say layers with hidden state (7 *7 *256). Could you let me know how to pass the…