Skip to content

PancakePrediction/Pancake_Pridction_KNN

Repository files navigation

image

Pancake_Pridction_KNN

For Pancake Pridction, Try to use the machine learning algorithm KNN to predict it.

Features

  1. You can follow the chart to quick bets.
  2. If you win the round, it will auto claim prize.
  3. If you lost the round, you can click the button +double to bet next round.
  4. You can collect more samples for the KNN, and Enable autobet to autoplay the prediction game.

Donations

This is a free project but any funding is appricated. ETH/BNB: 0x788c9F5406983Efe4f837d77CD7394Aca00Cb313

How to Enable KNN and collect samples automaticaly

  1. Create a MSSQL DB named StockData.
  2. Create a RoundData table in DB StockData with code bellow:
USE [StockData]
GO

SET ANSI_NULLS ON
GO

SET QUOTED_IDENTIFIER ON
GO

CREATE TABLE [dbo].[RoundData](
	[RoundID] [int] NOT NULL,
	[Last10sChange] [float] NULL,
	[LinkPriceSecounds] [int] NULL,
	[Offset] [float] NULL,
	[KDChange] [float] NULL,
	[Trend] [float] NULL,
	[KDRatio2] [float] NULL,
	[KDRatio3] [float] NULL,
	[Change] [float] NULL,
	[UpShadowLine] [float] NULL,
	[DownShadowLine] [float] NULL,
	[OnBollMiddle] [float] NULL,
	[IsBull] [bit] NULL,
	[ChangeBefore] [float] NULL,
	[K2] [float] NULL,
	[D2] [float] NULL,
	[K3] [float] NULL,
	[D3] [float] NULL,
	[BollWidth] [float] NULL,
	[BollChange] [float] NULL,
	[Change10m] [float] NULL,
	[Change15m] [float] NULL,
	[Change20m] [float] NULL,
	[Change25m] [float] NULL,
 CONSTRAINT [PK_RoundData1] PRIMARY KEY CLUSTERED 
(
	[RoundID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]

GO
  1. Create a RoundData_Prediction table in DB StockData with code bellow:
USE [StockData]
GO

SET ANSI_NULLS ON
GO

SET QUOTED_IDENTIFIER ON
GO

CREATE TABLE [dbo].[RoundData_Prediction](
	[RoundID] [int] NOT NULL,
	[PredictionIsBull] [bit] NULL,
	[PredictionIsBull2] [bit] NULL,
	[PredictionIsBullV2] [bit] NULL,
	[PredictionIsBullV3_0123] [bit] NULL,
	[PredictionIsBullV1_0123] [bit] NULL,
	[Score] [float] NULL,
	[TestSetAccuracy] [float] NULL,
	[kValue] [int] NULL,
	[AgainstPercent] [float] NULL,
 CONSTRAINT [PK_RoundData_Prediction1] PRIMARY KEY CLUSTERED 
(
	[RoundID] ASC
)WITH (PAD_INDEX  = OFF, STATISTICS_NORECOMPUTE  = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS  = ON, ALLOW_PAGE_LOCKS  = ON) ON [PRIMARY]
) ON [PRIMARY]

GO
  1. Fill your MSSQL connection info to config files.

Socials

Telegram: https://t.me/PancakePredictionKnn

Config

The Config is listed in Config\AppConfig.json like bellow:

{
  "faildCount": 7,					//Log for faild count.
  "betAmountInBNB": 0.1,				//init amount to bet in BNB.
  "AutoBet": false,					//Enable auto bet
  "Proxy": "http://127.0.0.1:7890",			//It can set to ""
  "DbServerName": "127.0.0.1,1433",			//If will enable KNN if you set this value, default is null or empty("")
  "dbName": "StockData",				//Db name for KNN, must be "StockData"
  "dbUser": "stockdataUser",				//Your db user
  "dbPassword": "stockdataUserPassword",		//Your db user's password
  "websocketNode": "wss://speedy-nodes-nyc.moralis.io/cde6a7978ca113c11c427bc5/bsc/mainnet/ws",		//If you want to use WS then set it otherwise leave it
  "wallletPrivateKey": "0x118e6050ccb1d2c03dbe71e340be74f4371a639afe6f0d1bbb9d288fdcf639d2",		//Your walllet private key
  "rpc_Endpoint": "https://bsc-dataseed.binance.org/"								//You can change it if you have better node
}

Recommand Bsc Node and Http Api

You can get a better node for quick bet and lessing data delay. https://ankr.com/ https://www.quicknode.com/

Running the project

If you want to run the project you can go to releases and a binary that will execute on your OS.

About KNN

  1. the K value 26 will be a good one.(Success 61.33%; Max Serial lost: 5, samples: 2000)
  2. More samples will make its predictions more accurate.

Warrning

  1. It's a gambling-like game, if you can't control your greed, you'll go broke.
  2. I went from 0.5 BNB to 7 BNB in 1 day through this game.
  3. But I went from 70 BNB to 0.01 BNB in half day.