-
Hi,
In the Caffe directory there is no instruction for IR ->Caffe conversion. Is this conversion supported?
I want to convert the weights in keras to caffe. How can I do this conversion using MMdn…
-
As we find in README.md in the top folder, we have some formats:
```
Exporting tags and assets to CNTK , Tensorflow (PascalVOC) or YOLO format for training an object detection model.
```
Then,…
-
I am converting a model from Caffe to CNTK, then from CNTK to ONNX. The Caffe model includes a dilated convolution, which is not supported on CNTK yet (at least not CPU). Is it supported by ONNX? Beca…
-
Since I upgraded docker-ce to 17.06, I can't prune stopped containers with name-based filters. (The same error happened with other kinds of filters).
The following code used to work:
`docker.from_…
-
I’d like to have a sequence of vectors output from bidirectional recurrent layer to be unpacked to a tensor. Then I’d like to apply cntk.times_transpose operation to that tensor, as in the following e…
-
sorry, I just can use cuda7.5, so I can only use cntk2.0 beta5. So the load model function is the old one loadmodel in cntk . can you give me an old format model file? Thanks
-
I want to retain the fast rcnn model which I have input training data A before. Now, I would like to add data B to optimize this model. So I modify A2_RunWithPyModel file,
[https://github.com/Micros…
-
By default the faster rcnn has a config for the AlexNet and VGG16 base models. I tried to get it to work with ResNet however I was not able to find the correct config parameters. I've used print para…
-
Platform (like ubuntu 16.04/win10):
Win 10
Python version:
Python 3.5.4
Source framework with version (like Tensorflow 1.4.1 with GPU):
TF 1.8.0 with GPU
Destination framework with version (like…
-
**Trying to create a custom loss function, which includes saving updated model weights, subprocess calls and numpy operations based on the updated weights after training of every batch. The loss funct…