User Portlet
Featured Contributor
| Discussions |
|---|
| Hi Jim, your complaint is perfectly valid, and I'm well aware we're missing a lot, not just relative to R or statsmodels, but to the old `ModelFit` as well. The good news is that most of what you're asking for is already on the roadmap for... |
| Great post! Thanks for sharing a concrete application =) Few tips. The net can be made both faster and more powerful by - embedding augmentation on the input image - using the stride to downsample - keep the convolution pyramid going deeper... |
| You mean something like this? ``` NetGraph[ AggregationLayer[Total, 1], "divide" -> ThreadingLayer[Divide, -1]|>, {NetPort["Input"] -> "norm", {NetPort["Input"], "norm"} -> "divide"} ] ``` |
| Hi Iuval, the third argument is used to extract training properties. To specify the probability cross-entropy instead of the index based one, you must use the `LossFunction` option. Try this: ``` famTrained = NetTrain[famNet,... |
| You can use the `LearningRateMultipliers` to freeze layers during training. Also, can you edit your post with the current solution? It will be useful for future people ending up here. Thanks! |
| The NN framework works on numerical data only. |
| The encoders and decoders are not layers but net port properties and you need to attached them to a port. This will work as you expect ``` dec = NetDecoder[{"Class", {"a", "b", "c"}}]; NNcompare = NetInitialize@ NetChain[{3,... |
| You could use "survived" -> NetDecoder[{"Function", First}] To pluck out the real from the list generated by `LinearLayer[1]` But the easiest would be to not attach anything to the "survived" port and just use LinearLayer[{}] ... |
| Hi Dalila, I cannot reproduce the error you see with this simple example ``` data = ResourceData["Sample Data: Fisher's Irises"]; c = Classify[data -> "Species"] FeatureImpactPlot[c] ``` Can you share a sample of the data you are using? |
| Hi Haoyu, the loss is just a function that is getting minimized during training. There is notthing special in the value 0. Typically, you can define it in a way that makes 0 mean "no error", "perfect result" or a similar concept but it's not... |