-
I can't see that the local gradients are ever reset. The values are overwritten by the global weights, but the optimizer `opt` is assigned to the global parameters, so won't this accumulate gradients …
-
We need to implement gradients of log partition functions for speed purposes. This is a relatively straightforward task that requires taking derivatives. It needs to be done for the entire list of dis…
-
At the moment the [`IntegratedGradients`](https://github.com/SeldonIO/alibi/blob/782ad53504d347f92c0ceb1cfbf5d1c376ceb677/alibi/explainers/integrated_gradients.py#L305) class only supports tf.keras an…
-
In Julia 1.7.3:
```
julia> Zygote.gradient(x -> x["foo"]["bar"]^2, Dict("foo" => Dict("bar" => 5)))
(Dict{Any, Any}("foo" => Dict{Any, Any}("bar" => 10)),)
julia> Diffractor.gradient(x -> x["f…
-
File "/root/miniconda3/lib/python3.8/site-packages/torch/cuda/amp/grad_scaler.py", line 212, in _unscale_grads_
raise ValueError("Attempting to unscale FP16 gradients.")
ValueError: Attempting t…
-
Gradients currently use initializer::ZEROS which initializes the gradient for the first iteration.
However, gradients are not initialized to zero for the next iterations and the layer is expected to …
-
Keras's implementation of Convolutional LSTM (https://www.tensorflow.org/api_docs/python/tf/keras/layers/ConvLSTM2D) uses `recurrent_intializer = orthogonal` to avoid exploding gradients.
This impl…
-
Should have some examples on page 1144 of pdf.pdf
-
Presently, it is very difficult to make a `FlxText` look distinct. I wish to improve `FlxText` to make it possible to have multiple outlines, multiple shadows, gradients, and perhaps other effects. Wh…
-
I clearly wasn't thinking about e-prop when I added support for training with validation in #57 as currently gradients are being accumulated e.g. at https://github.com/genn-team/ml_genn/blob/master/ml…