yydscoder

I have just been unbanned due to an error on fraud squads part and all of my devlogs are gone now lol? Thanks a lot fraud squad. I cant even ship an update to the project because the timeline is disorted now.

Added DOT, as well as rewrote the healing logic of how heals work since i feel like they clog up the hand a bit. In the screenshot you can see that there is a constant tick * number of turns on the enemy. This will be the gimmick of the nature/earth deck of which i have added 15 cards themed around healing and DOT and out sustaining the enemy with heals. I have also overwritten the logic around how I want the game to progress but its still in the testing phase although it can be demoed against the static slime enemy. I plan to add more elemnents and themes then think about game progression in the coming few sessions.

Added around 30 new water cards. I feel really tired right now. I still have to do earth elements and wind later, as well as develop enemies and game mechanics, but a demo can be released soon. Since voting is disabled and I have a negative vote balance, I cannot ship it as a demo, which is tragic. This time can be used for more updates and development, I suppose. In this devlog, I will be showing the water class. As you can see, there is a fire spell there, so I am unsure if I should support mixing elements for cards. Maybe a deckbuilder could come, and I could have levels to make it something similar to an RPG.

Ah yes, I also made it so that spells you have sufficient mana for will glow. I probably need to debug a few cards that are mainly utility cards, such as “Phoenix” and “Regen,” since they clog up your hand rotation. But yes, so far it has been a very productive 9 hours of work. I have also swapped the player and enemy, and will have to change the positioning of the bars and also change how the healthbars are positioned and maybe will put in new graphics for the turn element.

I have added another ~15 cards under the “fire element” for greater diversity to add onto the game engine, the fire region will mostly encompass the “Sacrifice HP” for progress playstyle. I have also been working on the game mechanics, mostly in relation to Mana and how it interacts with the players progression. So far I only have the slime enemies but i intend on making more and more enemies to extend the length of the game. I have also been thinking about adding basic lore behind these cards perhaps since I kinda want this project to develop in the future. I have added logging as well as an end turn but no enemy reaction yet. But so far I think my emoji approach is working pretty well since I do not have to draw cards. Each card is about 200 to 300 lines of code which is really tiring to code but I suppose it does help that I am clocking in these hours. Cant wait to release a prototype!

Okay so of course im using emojis for my characters because i dont want to have to dreaw them, it can be work done in the future. I have also added victory and defeat screens and the basic engine of the entire game. I will be adding extra cards and different themed decks in the future. I have coded a victory screen and a defeat screen but i have not added the conditions for defeat. I probabyl also need to add more game mechanics but that will come in the future. I have also built SaveSystem.js: Save wins/losses to local storage for the future as you can see in the last screen debug, there is a log of a streak of 6 no matter how many times you reload and increases by 1 every time you have a streak!

Today I have built health bars which as you can see attached are done via CSS, and i also added css animations for damage and screen shaking effects which i unfortunately cant show here, as well as crit animations. I have also added deeper debugging as you can see. So far the key engine is going great and I am having lots of fun!

Day 3 was about building the enemy and combat logic for my game. I created three new files: Enemy.js, SlimeEnemy.js, and DamageCalculator.js. First, I made the main Enemy class with HP, attack, and an emoji so it can show up visually in the game. Then I made a SlimeEnemy that extends the base enemy class, so I can easily create different types of enemies later. After that, I worked on DamageCalculator.js, where I wrote the math and logic for calculating damage that i may change in the future, including defense and critical hit ratios. The most important part today was connecting everything together so when I click a card, it deals damage to the enemy and updates their HP on the screen. If the HP reaches 0, the enemy dies properly. I also wrote unit tests in src/tests/combat.test.js to test the combat logic, As you can see the slime here gets defeated and a little popup will greet you !

Due to the updates, i was suggested to ship it again since the payout was quite mediocre

AAAAAAAAAAAAAAAAAAAAAAAA I FINALLY GOT IT WORKING, I ADDED MY FIRST CARD AND ALSO HAD A LITTLE LIGHT UP ANIMATION. For the project I will be sticking to emojis since I do not have time for graphic illustrations of cards and they also dont contribute to hours done sooooo

I FORGOT TO DEVLOG!!!!!!!!!!!!!!!! I was too lost in the sauce!!

Heres what i did

Built engine.js: Game loop , turn manager.
Built state.js: Player HP, Enemy HP, Mana, Turn Count.

I have also built the basic stage where you will fight your opponent, maybe I will make it top down instead of left to right and have a more hearthstone themed approach. MORE TO COME!

I think we can look to ship soon, i have added a rules based fallback and a way for the model to know if something follows the 5-7-5 structure and to provide a fallback if it doesnt conform to it. Now I just need to rewrite the readme and we can look to ship

Wtf is my model doing, was training for 4 ish hours and i got this haiku generated which is extremely obscene, will need to retry and retrain the model and put in guardrails against this type of content

Tuning for windows and making it easier for the user to download externally required modules for pytorch such as windows visual c++

Finally finished model training, now I will download it and progress with local testing
Local testing has been done but the model is incoherent at times, will need to refine it further, maybe i can integrate it with hackai endpoint to refine it based on the output of the local model

Today was a hassle, most of the work was done on kaggle where I have started the model training and will HOPEFULLY be done by tommorow. Its a shame that the configuration on kaggle does not contribute to the time for this project

Alright I am starting this project from scratch, i am currently writing and importing datasets from kaggle regarding poetry from poetry foundation and gpt 2 as my lightweight model, but i am a bit confused as to whether it would be better to attach my dataset that is trained onto git or to have it hosted on hugging face and connected via an api call.