Update documentation.

Author: JoshuaBillson

docs/src/index.md

@@ -32,7 +32,7 @@ We can fit our model to our data by calling `fit!(m, X, Y; opt=Flux.Adam(), batc
 containing the learning curve, best epoch, and lowest loss observed during training as its second value. We can use Makie's `lines` method to visualize
 the learning curve.
 ```julia
-model, report = tMixtureDensityNetworks.fit!(model, X, Y; epochs=500, opt=Flux.Adam(1e-3), batchsize=128)
+model, report = MixtureDensityNetworks.fit!(model, X, Y; epochs=500, opt=Flux.Adam(1e-3), batchsize=128)
 fig, _, _ = lines(1:500, lc, axis=(;xlabel="Epochs", ylabel="Loss"))
 ```
 
@@ -49,7 +49,7 @@ axislegend(ax, position=:lt)
 
 ![](figures/PredictedDistribution.png)
 
-We can also visualize the conditional distribution predicted by our model at x = -2.0.
+We can also visualize the conditional distribution predicted by our model at x = -2.1.
 ```julia
 cond = model(reshape([-2.1], (1,1)))[1]
 fig = Figure(resolution=(1000, 500))

src/layers.jl

@@ -1,7 +1,7 @@
 """
 $(TYPEDEF)
 
-A layer which returns a univriate Gaussian Mixture Model as its output.
+A layer which produces a univariate Gaussian Mixture Model as its output.
 
 # Parameters
 $(TYPEDFIELDS)
@@ -20,7 +20,7 @@ $(TYPEDSIGNATURES)
 Construct a layer which returns a univariate Gaussian Mixture Model as its output.
 
 # Parameters
-- `input`: Specifies the size of the input tensor, which should have the dimensions `input x N`.
+- `input`: Specifies the length of the feature vectors. The layer expects a matrix with the dimensions `input x N` as input.
 - `mixtures`: The number of mixtures to use in the GMM.
 """
 function UnivariateGMM(input::Int, mixtures::Int)
@@ -53,7 +53,7 @@ end
 """
 $(TYPEDEF)
 
-A layer which returns a multivariate Gaussian Mixture Model as its output.
+A layer which produces a multivariate Gaussian Mixture Model as its output.
 
 # Parameters
 $(TYPEDFIELDS)
@@ -74,8 +74,8 @@ $(TYPEDSIGNATURES)
 Construct a layer which returns a multivariate Gaussian Mixture Model as its output.
 
 # Parameters
-- `input`: Specifies the size of the input tensor, which should have the dimensions `input x N`.
-- `output`: Specifies the dimension of the labels, which should have the dimensions `output x N`.
+- `input`: Specifies the length of the feature vectors. The layer expects a matrix with the dimensions `input x N` as input.
+- `output`: Specifies the length of the label vectors. The layer returns a matrix with dimensions `output x N` as output.
 - `mixtures`: The number of mixtures to use in the GMM.
 """
 function MultivariateGMM(input::Int, output::Int, mixtures::Int)

src/model.jl

@@ -16,12 +16,13 @@ Flux.@functor MixtureDensityNetwork (hidden, output)
 """
 $(TYPEDSIGNATURES)
 
-Construct a new MixtureDensityNetwork.
+Construct a standard Mixture Density Network.
 
 # Parameters
-- `input`: The dimension of the input features.
+- `input`: The length of the input feature vectors.
+- `output`: The length of the output feature vectors.
 - `layers`: The topolgy of the hidden layers, starting from the first layer.
-- `mixtures`: The number of Gaussian mixtures to use in the conditional distribution.
+- `mixtures`: The number of Gaussian mixtures to use in the predicted distribution.
 """
 function MixtureDensityNetwork(input::Int, output::Int, layers::Vector{Int}, mixtures::Int)
     # Define Weight Initializer