cross-posted from: https://lemmy.world/post/28461880

so I spent last several days making collecting watch time on both videos and livestreams more robust and work across multiple peertube instances, im sure it still has gaps in the structure so that jenk data can get in.

if you want to try it heres the link https://github.com/solidheron/peertube_recomendation_algorythm/ btw its a browser extension

so now I got two parts left that I know of first being creating the user_recomendation_vector and the function that gets recommendation based on that vector. I settle on cosine similarity vector since its easy to implement and can be run in browser with only data collected by the user device, and doesnt requires sharing outside of peertube api. user_recomendation_vector should have two part AOLR: (algorithm of last resort) which will be the words in the title, tags, and description tokenized with an float value and recomended_standard: which will be based on what category eit

so I spent last several days making collecting watch time on both videos and livestreams more robust and work across multiple peertube instances, im sure it still has gaps in the structure so that jenk data can get in.

if you want to try it heres the link https://github.com/solidheron/peertube_recomendation_algorythm/ btw its a browser extension

so now I got two parts left that I know of first being creating the user_recomendation_vector and the function that gets recommendation based on that vector. I settle on cosine similarity vector since its easy to implement and can be run in browser with only data collected by the user device, and doesnt requires sharing outside of peertube api. user_recomendation_vector should have two part AOLR: (algorithm of last resort) which will be the words in the title, tags, and description tokenized with an float value and recomended_standard: which will be based on what category either programs or people decide a video belongs to along with an asso

cross-posted from: https://lemmy.world/post/28241019

working on peertube recommendation algorithm has made me realize that we should be collecting our own data.

my program that monitors peertube watching habit and stores it locally and should track YouTube watching habit just so that data can be used in a recommendation algorithm. it would be a good idea for fediverse if there were programs that monitors and stores the data locally for the fediverse and corporate counterpart.

the reason why I would track youtube data was to get the titles of the video

working on peertube recommendation algorithm has made me realize that we should be collecting our own data.

my program that monitors peertube watching habit and stores it locally and should track YouTube watching habit just so that data can be used in a recommendation algorithm. it would be a good idea for fediverse if there were programs that monitors and stores the data locally for the fediverse and corporate counterpart.

the reason why I would track youtube data was to get the titles of the video

cross-posted from: https://lemmy.world/post/28197750

I made a browser extension that track the video you watch on a peertube instance linked here https://github.com/solidheron/peertube_recomendation_algorythm

it works decently well but it still has some glitches and a bug that causes minor information loss that will need to be patched out. so this is in alpha and needs to be worked on. I do have another script that gets the data of a hundreds of peertube videos and the closed captions. so got some basic components of an algorithm started. next ill probably figure out how to categorize videos based on peertube api and closed captions.

It's weird knowing someone will improve on this at some point

bellow is the data structure of the browser extension. I did watch videos to get that data and all of this will be stored locally. I recommend looking at it with https://jsonformatter.org/json-viewer . leave a comment/suggestion, I'm up for any submissions.

[{"title":"The

I made a browser extension that track the video you watch on a peertube instance linked here https://github.com/solidheron/peertube_recomendation_algorythm

it works decently well but it still has some glitches and a bug that causes minor information loss that will need to be patched out. so this is in alpha and needs to be worked on. I do have another script that gets the data of a hundreds of peertube videos and the closed captions. so got some basic components of an algorithm started. next ill probably figure out how to categorize videos based on peertube api and closed captions.

It's weird knowing someone will improve on this at some point

bellow is the data structure of the browser extension. I did watch videos to get that data and all of this will be stored locally. I recommend looking at it with https://jsonformatter.org/json-viewer . leave a comment/suggestion, I'm up for any submissions.

[{"title":"The Blues Brothers (1980) - Dalek Zone","url":"https://dalek.zone/w/56ozwq8e

cross-posted from: https://lemmy.world/post/28140813

Little update on making a peertube recommendation algorithm. I made a script that collects video json and closed caption from instances. Later on I'll figure out a way to organize the data into video description vector. (Data manipulation will be ever improving process but data collecting should be stagnant and only change with activityhub and peertube updates)

My question is if there's a program that monitors or gets a users video watch time and interaction data and stores it locally. I just need something to record my video interactions of peertube videos such as like or dislike a video.

Little update on making a peertube recommendation algorithm. I made a script that collects video json and closed caption from instances. Later on I'll figure out a way to organize the data into video description vector. (Data manipulation will be ever improving process but data collecting should be stagnant and only change with activityhub and peertube updates)

My question is if there's a program that monitors or gets a users video watch time and interaction data and stores it locally. I just need something to record my video interactions of peertube videos such as like or dislike a video.

cross-posted from: https://lemmy.world/post/28064839

writing down my thoughts for added elements called Video_description_vector and Video_description_vector_history. Video_description_vector is an element that an instance gives to describe what categories a video does and does not belong to and Video_description_vector_history is record of user submissions of what categories they say a video belongs to so that way categories can be removed.

Video_description_vector has sub elements that are format standards, this is done so that so that future potentially better format can be entered into Video_description_vector . I'm working on my recommended standard, so far I have isTrue array that lists what categories the video belongs to, while the isFalse element lists what categories it does not belong to—subjectively. done that way so that isn't a sea of element like "cartoon":True or "action":False . I do know that I need sub-element for music metrics

Below is a made-up exampl

writing down my thoughts for added elements called Video_description_vector and Video_description_vector_history. Video_description_vector is an element that an instance gives to describe what categories a video does and does not belong to and Video_description_vector_history is record of user submissions of what categories they say a video belongs to so that way categories can be removed.

Video_description_vector has sub elements that are format standards, this is done so that so that future potentially better format can be entered into Video_description_vector . I'm working on my recommended standard, so far I have isTrue array that lists what categories the video belongs to, while the isFalse element lists what categories it does not belong to—subjectively. done that way so that isn't a sea of element like "cartoon":True or "action":False . I do know that I need sub-element for music metrics

Below is a made-up example. It’s about the Cleveland Browns, an American football team, do

cross-posted from: https://lemmy.world/post/28011368

So I started by doing research and by research I mean watching two videos on YouTube about basic recommendation algorithms.

I did watch a 30 minute video on Netflix software engineer talking about using machine learning and complex matrix and these bandit style machine learning algorithms to recommend TV shows/movies really the base conclusion is that there's a 50% improve over doing all these complex things over their baseline measurement. Baseline will mean traditional pre neral network based algorithms.

The way I interpret it is that basics take you a long way and all the basics are is just organizing any peertube video into a vector and people watching into a vector as well. The idea would be that which videos are more similar to each other would be good recommendations if a watcher watch one of those videos, or if they didn't like it don't recommend any videos similar to that. Once these videos get vectorized then t

So I started by doing research and by research I mean watching two videos on YouTube about basic recommendation algorithms.

I did watch a 30 minute video on Netflix software engineer talking about using machine learning and complex matrix and these bandit style machine learning algorithms to recommend TV shows/movies really the base conclusion is that there's a 50% improve over doing all these complex things over their baseline measurement. Baseline will mean traditional pre neral network based algorithms.

The way I interpret it is that basics take you a long way and all the basics are is just organizing any peertube video into a vector and people watching into a vector as well. The idea would be that which videos are more similar to each other would be good recommendations if a watcher watch one of those videos, or if they didn't like it don't recommend any videos similar to that. Once these videos get vectorized then the watcher's vector can be updated in a basic way more watch tim

so this is code that the owncast plays https://github.com/solidheron/owncast_plays_dualshock script is based on, but the peertube version. if you were interested in the script here a trial run of the script. with the peertube setting and obs studio it should be similar latency chatter experiences with owncast

should note that for some reason the DiskStation emulator is not registering r2 and l2 button presses on linux. I don't think the game has any use for those buttons

rotated out the games and script. don't worry dragon quests monster has a save state and can be loaded back at any time

rotated out the games and script. don't worry dragon quests monster has a save state and can be loaded back at any time

I finished up the script that controls the PS4 controller and it works good enough with one know issue. I look at the twitch plays Pokemon channel and they rotate out there games about every 10 days so I'll take down dragon quest monsters to free it up for chocobo dungeons 2. Dragon quest will be returned to on a later date.

It's just that activity on dragon quest fedi plays has been sparse.

Finally converted my "PeerTube Plays" script to work with Owncast! Repo here: https://github.com/solidheron/owncast_plays_dualshock

The script reads messages from an Owncast chat URL—example: https://live.24hrkpop.com/embed/chat/readwrite/%E2%80%94and maps chat messages to PlayStation 4 controller inputs. It doesn’t control the analog sticks, but it does handle every other button, including the touchpad.

Basic usage:

Typing up presses up on the D-pad

Typing square presses the square button

And my favorite: x o 15 doesn’t send hugs and kisses—it presses both cross and circle 15 times in a row

You can't press opposites dpad presses at once (e.g., up + down or left + right), but everything else is fair game.

As I mentioned in my last post: Put it in player 2’s slot, use it during an intermission screen, or overlay it on top of a streamer’s real inputs for some chaotic sabotage fun.

I made another script for my fedi plays stream linked here https://github.com/solidheron/peertube_chat_controller that script is only for peertube chat and it's used to control a virtual ps4 controller that you can connect to anything usually emulators.

I should able to convert make an Owncast version, but I do have questions about if the owncast version of this script should open a web browser and scrape the comments or if owncast hosts have different access to the comments on their stream's comment section.

it should be fun if chat had a virtual controller then they can be player 2 (little sibling streams), share input with the streamer (punishment challenge/sabotage stream), or have a intermission scene gimic where chat plays a game

I keep twerking the script hopefully it doesn't execute a previous chat anymore. The script won't parse any data until a new message appears in chat. Also I lowered new messages check by 2 seconds so latency is reduced by 2 seconds and commands can be stringed faster.

I have looked at old VODs of twitch plays Pokemon the one involving final fantasy 7 and the commands get put onto a stack and executed in chronological order. These vods are 4 years ago and the original design idea was based on twitch plays Pokemon red that seemed to execute whatever is most recent

Now I'm working on a script that takes peertube chat and puts them into inputs on a virtual PS4 controller. So far it works but it needs a way to do two button presses

I might make a fedi plays for chocobo dungeon on the ps1

turns out that ety has edited VODs