Morgensol

Results 9 comments of Morgensol

Convert the model (checkpoints) to .pb and .pbtxt for OpenCV use

> Hi, > > I managed to train a model with my own data. Now I want to use the model in OpenCV. OpenCV requires both .pb and .pbtxt as...

Invalid argument: indices[3824] = [10, 1, 52, 0] does not index into shape [16,52,52,3,6] while training on xView Dataset 1 class

That can be several things, but it does really sound like your box labels might be larger than 1 or smaller than 0, which is a problem i have run...

Invalid argument: indices[3824] = [10, 1, 52, 0] does not index into shape [16,52,52,3,6] while training on xView Dataset 1 class

> Are there more steps than just changing "num_classes" to be 1? if there are, could you send me a link to it? Thank you so much. Yes, if you...

Invalid argument: indices[3824] = [10, 1, 52, 0] does not index into shape [16,52,52,3,6] while training on xView Dataset 1 class

> \ --classes ./data/sat.names What is the content of your sat.names?

Invalid argument: indices[3824] = [10, 1, 52, 0] does not index into shape [16,52,52,3,6] while training on xView Dataset 1 class

Could you post a sample of one of your labels, perhaps that could give me some insight. If i were you id check all your labels to 100% certain that...

Invalid argument: indices[3824] = [10, 1, 52, 0] does not index into shape [16,52,52,3,6] while training on xView Dataset 1 class

Not that its an issue, but it does seem like in this labelfile that almost all of your labels are en the bottom left corner of your image. I dont...

The training loop shows one extra dimension of loss.

Hi, This has nothing to do with the amount of classes in your dataset, this is simply due to how yolo tiny is built compared to normal yolo. The reason...

tensorflow.python.framework.errors_impl.InvalidArgumentError: Received a label value of -1 which is outside the valid range of [0, 80).

What .names file are you using for classes? What weights? Need more info, it says that its recieving a -1 label, which usually means that the label you are passing...

Expose "require" suite

> [@Blquinn](https://github.com/Blquinn), if you choose (require/assert) at compile time then you can do the following (per struct, or as a base to other test structs): > > type Suite struct...