Added Summary function for model summary

src/utils.jl

@@ -200,3 +200,76 @@ be used for aesthetic purposes, or by recovering Python users.
 macro jit(ex)
   esc(ex)
 end
+
+############################################################################
+    # Summary of the model # 
+############################################################################
+struct Trainable_layer
+    layer_name::Any
+    num_params::Int64
+end
+
+struct Summary
+    layers::Array{Trainable_layer}
+    total_params::Int64
+end
+
+"""
+    print_summary(Summary)
+Helper function for printing the summary of the model in a proper format. 
+"""
+function print_summary(summ)
+
+    print("+++++++++++++++++++++++++++++Model Summary++++++++++++++++++++++++++++++++++++\n")
+    for i in 1:length(summ.layers)
+        layer = summ.layers[i]
+        print("")
+        print("Trainable Layer_$i: ",layer.layer_name,"     ")
+        print("Num_Trainable_Parameters: ",layer.num_params,"\n\n")
+    end
+    print("Total Number of Trainable Parameters: ",summ.total_params,"\n")
+end
+
+
+""""
+    Summary(model)
+Call `Summary(model)` to get the brief summary of the model. It return struct of two fields: layers and total_params.
+
+Fields/Parameters:
+
+1.`Summary(model).layers`: gives a list of trainable layers in the model, along with the number of parameters in each layer.
+    It returns an array of struct `Trainable_layers`, which have fields `layer_name` and `num_params`.<br>
+<br>
+2.`Summary(model).total_params`: returns the total number of trainable parameters used in the model
+
+```julia
+#Example
+julia> model = Chain(Dense(10,64), Dense(64,1))
+Chain(Dense(10, 64), Dense(64, 1))
+
+julia> Flux.Summary(model);
++++++++++++++++++++++++++++++Model Summary++++++++++++++++++++++++++++++++++++
+Trainable Layer_1: Dense(10, 64)     Num_Trainable_Parameters: 704
+
+Trainable Layer_2: Dense(64, 1)     Num_Trainable_Parameters: 65
+
+Total Number of Trainable Parameters: 769
+```
+"""
+function Summary(model)
+    layers_vec =[]
+    layers = model.layers    
+    Total_trainable_par =0
+    for layer in layers
+        try
+          x = sum(length,params(layer))
+          push!(layers_vec,Trainable_layer(layer,x))
+          Total_trainable_par+=x
+        catch error
+        end
+    end
+    Summ = Summary(layers_vec,Total_trainable_par)
+    print_summary(Summ)
+    return Summ
+end
+#######################################################################################