spherical | Display 360x180 panoramas with HTML5

This is my first post here and I am not that experienced, so please excuse my ignorance.

I am building a Monte Carlo simulation in C++ for my PhD and I need help in optimizing its computational time and performance. I have a 3d cube repeated in each coordinate as a simulation volume and inside every cube magnetic particles are generated in clusters. Then, in the central cube a loop of protons are created and move and at each step calculate the total magnetic field from all the particles (among other things) that they feel.

At this moment I define everything inside the main function and because I need the position of the particles for my calculations (I calculate the distance between the particles during their placement and also during the proton movement), I store them in dynamic arrays. I haven't used any class or function,yet. This makes my simulations really slow because I have to use eventually millions of particles and thousands of protons. Even with hundreds it needs days. Also I use a lot of for and while loops and reading/writing to .dat files.

I really need your help. I have spent weeks trying to optimize my code and my project is behind schedule. Do you have any suggestion? I need the arrays to store the position of the particles .Do you think classes or functions would be more efficient? Any advice in general is helpful. Sorry if that was too long but I am desperate...

Ok, I edited my original post and I share my full script. I hope this will give you some insight regarding my simulation. Thank you.

Additionally I add the two input files

parametersDiffusion_spher_shel.txt

parametersIONP_spher_shel.txt

https://codesandbox.io/s/basic-demo-forked-yup2o?file=/src/App.js

I have the above sandbox of a spherical word cloud consisting of random words - I'm trying to modify the code so rather than random words, the cloud can display words from a list of my choosing, for example

let MyList = ['React', 'Node', 'SQL', 'NoSQL', 'TDD', 'JavaScript', 'Python', 'Git', 'Excel', 'ReactNative', 'HTML', 'CSS', 'TypeScript', 'Java', 'Angular', 'Django']

I tried doing this by editing the code inside the for loop of the cloud function by changing randomWord() to wordFromMyList() which is defined below.

I took an equirectangular envMap from the three.js docs sample where it appears sharp on a sphere.

If I load the same material into my aframe scene with the following params

I am using the Rails Geocoder gem to geocode the latitude and longitude of a street address submitted by a user. I would also like to automatically pull a users IP location to populate a home screen with other user submitted addresses near their location.

I'd like to use an API key for IPinfo to get the user's location and a different API key for Google to geocode the lat/lng of the street address. Previously, when I call out two api keys I get a warning message stating the first key was overwritten by the second, probably because they had the same name.

I saw this previous post where a user answered saying they can do it by passing the service as a lookup value but how did they set up the api key in the geocoder.rb config file? Did they change the api name for the second service? Did they have to add in another config file?

Any help at all would be appreciated. My current config file is below. I commented out the IPinfo lookup for now so I can still geocode items in production. Thank you.

I am trying to use the KMeans clustering from faiss on a human pose dataset of body joints. I have 16 body parts so a dimension of 32. The joints are scaled in a range between 0 and 1. My dataset consists of ~ 900.000 instances. As mentioned by faiss (faiss_FAQ):

As a rule of thumb there is no consistent improvement of the k-means quantizer beyond 20 iterations and 1000 * k training points

Applying this to my problem I randomly select 50000 instances for training. As I want to check for a number of clusters k between 1 and 30.

Now to my "problem":

The inertia is increasing directly as the number of cluster increases (n_cluster on the x-axis):

I tried varying the number of iterations, the number of redos, verbose and spherical, but the results stay the same or get worse. I do not think that it is a problem of my implementation; I tested it on a small example with 2D data and very clear clusters and it worked.

Is it that the data is just bad clustered or is there another problem/mistake I have missed? Maybe the scaling of the values between 0 and 1? Should I try another approach?

Contain the necessary files. Add this to your "My Drive". https://drive.google.com/drive/folders/1epROVNfvO10Ksy8CwJQdamSK96JZnWW9?usp=sharing Google colab link with a minimal example: https://colab.research.google.com/drive/18sMqNn8IpTQLZBlInWSbX0ITd2GWDDkz?usp=sharing

This basic block 'module', if you will, is part of a larger network. It all boils down to this block, however, as this is where the convolutions are performed (in this case depthwise separable convolution). The network is seemingly able to train, however, while training (and during all the epochs) a WARNING is thrown out:

I'm installing a remote network of spherical/360-degree cameras and need to have the images they produce all aligned with something (north). The problem is, the cameras might end up being mounted facing any random direction, so I need to shift/offset (horizontally) the image, and wrap whatever ends up outside the bounds of the image to the other side.

After taking a sample image after installation of each camera, I will know the correct pixel value to offset the image and will be able to specify the offset value on a per-camera basis.

The first half of this article explains how to do this in Photoshop (Filter menu > Other > Offset, specify an offset pixel value, and then enable the "wrap around" option). Photoshop offsets the image by the defined number of pixels, and any excess is automatically moved to the other side of the image to fill in the gap created by shifting the image.

I just need to automate the offsetting of the image and wrap the excess around to the other side of the image so that no part of the image is actually lost, in a Linux (Raspbian) environment, via the command line. I am not able to use a GUI to do this as we're talking about millions of images in the long-term.

I tried doing this with Imagemagick's geometry offset tool (convert -geometry +0+100 R0011996.JPG), but a) I couldn't figure out how to actually use it, and b) there doesn't seem to be an option like the "wrap around" option in Photoshop.

I just realised The GIMP has an equivalent tool and a command-line interface and tried that, but couldn't find any decent documentation. I found the Procedure Browser, which said to use "gimp-drawable-offset". I tried the following, but it failed:

I'm trying to generate (as efficiently as possible), a random uniform, 3D distribution of points inside of a sphere of an inner radius r_min and outer radius r_max,i.e. a shell. I found a similar solution here: Sampling uniformly distributed random points inside a spherical volume, however this is only true for a whole sphere of r_min=0 and r_max=1. This is done using the following code:

Is it possible to add a satellite view to the OpenLayers?

This is my method that initializes the OpenLayers: