FX price prediction using state-of-the-art image recognition Machine Learning algorithms Experiment to see if an accurate prediction of long/short positions in FX can be done using state-of-the-art image recognition algorithms (via fast.ai library). How does it work - Get the data you need in a pandas DataFrame - Manipulate the data according to your needs - Plot the data using matplotlib - Save the plots as pictures with appropriate labels and placed in the appropriate data folder for later use - Run the prediction model - Be dissapointed with the prediction accuracy and try again :) 1. OandaDataRetrieval This is the initial step of data preparation, plotting and exporting as images. Note that even though i'm using this example for FX data (using Oanda as a broker), with minimal code adjustment you can use your own data sources/DataFrames Data retrieval and preparation - Data is being pulled via Oanda's API for whatever time window you require. Start (e.g 2018-01-01) and End dates can be specified, instrument (e.g GBPUSD) as well as candle freqency (e.d M1 meaning 1 minute candles).

⚠ No verified equity curve — no track-record source connected.

No score history is stored yet — only the current score is shown.

⚠ No live verification account connected — ask for proof before buying.

Alerts on changes: coming soon

Prop-firm compatibility not provided.

Data-completeness & trust index (not a profitability rating)