sclbl_101_model

sclbl-101-tutorial.py

# Imports
import numpy as np
from sklearn.linear_model import LinearRegression

# Data generation y = 10 + 2*x1 -.5*x2 + noise:
np.random.seed(0) 
n = 100
x1 = np.random.uniform(0,10,(n,))
x2 = np.random.uniform(0,10,(n,))
y = 10 + 2 * x1 -.5*x2 + np.random.normal(0,1,(n,))
X = np.column_stack((x1, x2))

# Model fitting
lm = LinearRegression()  
lm.fit(X, y.reshape(-1, 1))


# Scailable demo:
import sclblpy as sp

# Create an example feature vector
fv = np.array([2,5])  

# Create documentation for this model (accepts Markdown)
docs = {}
docs['name'] = "Simple linear regression demo"
docs['documentation'] = """#Linear regression demonstration.
	\nFor the [getting started tutorial](https://github.com/scailable/sclbl-tutorials/blob/master/README.md)."""

# Upload the model to transpile to WASM and make available
sp.upload(lm, fv, docs=docs)
              
# Note: the last call will only work if you have a valid scailable account
#   get one at https://admin.sclbl.net/signup.html
#   and install sclblpy using pip.