louisoutin/deep_gas_oracle
Unofficial implementation of the paper: "Step on the Gas? A Better Approach for Recommending the Ethereum Gas Price"
Implements a GRU-based neural network that forecasts Ethereum minimum gas prices over 3-hour windows using 5-minute resampled historical block data from BigQuery, then derives recommendations by analyzing prediction slopes. The model trains on historical gas price and blockchain data from Kaggle and Google BigQuery, with evaluation notebooks comparing performance against Geth's default gas estimation strategy.
No commits in the last 6 months.
Stars
23
Forks
6
Language
Jupyter Notebook
License
MIT
Category
Last pushed
Jan 27, 2021
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/louisoutin/deep_gas_oracle"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related frameworks
mesjou/price_simulator
simulation platform for algorithmic pricing
MohammadvHossein/predict-gold-price
A Python project that forecasts gold prices using machine learning and deep learning techniques,...
VMS-Solutions/predict_leadtime
Prediction of Copper Supply Lead Time Based on Crude Oil and Raw Material Prices
ramtiin/Predicting-Stock-Prices-Using-FB-Prophet
Prophet is a procedure for forecasting time series data based on an additive model where...
yennhi95zz/predict-gold-prices
Gold price forecasting using time series is a statistical technique that involves analyzing...