linear_regression 🚀 
Machine learning introduction
Hypothesis 🚕
Ro = b + m * o
Gradient descent 🚠
tmp0 = p0 - lrate * (d * sum(hyp - y))
tmp1 = p1 - lrate * (d * sum((hyp - y) * X))
p0 = tmp0
p1 = tmp1
Implementation ✈️
A trainer which train a model given an datasets
The dataset must follow this pattern:
columns: [km, price]
A reader which predict price based on a model
The model must follow this pattern:
columns: [theta0, theta1, Xmin, Xmax, ymin, ymax]
Usage ⛵
python train.py
usage: python train.py [-h] [--path path] [--out path] [--plot PLOT]
[--epochs EPOCHS] [--lrate LRATE]
Train a model given a dataset
optional arguments:
-h, --help show this help message and exit
--path path must be a valid dataset path
--out path must be a valid output path
--plot PLOT plot training
--epochs EPOCHS number of iteration
--lrate LRATE number of iteration
python predict.py
usage: predict.py [-h] [--path path] [-p path] <mileage>
Predict the price given an mileage
positional arguments:
<mileage> must be a valid integer
optional arguments:
-h, --help show this help message and exit
--path path must be a valid model.csv path
-p path must be a valid model.csv path