shapes | angry rotating shapes | Frontend Framework library

Say I have an MLP that looks like:

I am trying to make a next-word prediction model with LSTM + Mixture Density Network Based on this implementation(https://www.katnoria.com/mdn/).

Input: 300-dimensional word vectors*window size(5) and 21-dimensional array(c) representing topic distribution of the document, used to train hidden initial states.

Output: mixing coefficient*num_gaussians, variance*num_gaussians, mean*num_gaussians*300(vector size)

x.shape, y.shape, c.shape with an experimental 161 obserbations gives me such:

(TensorShape([161, 5, 300]), TensorShape([161, 300]), TensorShape([161, 21]))

I am currently a bit stuck! Lets say, have a grid of shapes (nested For-Loop) and I want to use a wave to animate it. The wave should have an offset. So far, i can achieve it. Currently the offset affects the Y-axis … But how can I manage to have a RADIAL offset – you know – like the clock hand, or a radar line… I really would like the offset to start from (width/2, height/2) – and then walks around clockwise. Here is my code and the point where I am stuck:

I have tried all possible ways to implement an inner border radius but I can't implement this. Have you anyone know about this.

I have referred this but I can't draw this layout.

I'm trying to automate network diagrams and I'm having trouble getting rid of the label of the cloud shape. When I try to get rid of the -Label parameter, the cloud will not be drawn. I know that I can manually delete the label but is there a way to draw the cloud without using the -Label parameter? I've provided my code down below:

I'm struggling with the following issue. I visualized a big social network and would like to customize the color palette of the edges captured in geom_segmentfor better visibility. For instance, I want to replace my blue scale by a red scale. How can I achieve that in the easiest way possible?

I have the following code which is working fine:

I am trying to build a Keras model for a classification model and I get and error while I am trying to fit the data.

ValueError: Shapes (None, 99) and (None, 2) are incompatible

Code:

I've tried for the last few days without too much success to rotate, scale and translate shapes on the canvas. I've read everything I could find on internet about similar issues but still I cannot seem to be able to adapt it to my own problem.

If everything is drawn on the same scale, I can still drag and drop. If I rotate the shapes, then the mouseOver is messed up since the world coordinates don't correspond anymore with the shape coordinates. If I scale, then it's impossible to select any shape. I look at my code and do not understand what I'm doing wrong.

I read some really nice and detailed stackoverflow solutions to similar problems. For example, user @blindman67 made a suggestion of using a setTransform helper and a getMouseLocal helper for getting the coordinates of the mouse relative to the transformed shape.

inverse transform matrix

I spent some time with that and could not fix my issue. Here is an example of what I tried. Any suggestion is appreciated.

I'm trying to iterate over a List of objects, for each of the object I want to display a composable card. The problem is that you cant call Composable functions from inside the list.forEach{} brackets.

The Code: