Prratek Ramchandani prratek

Principal Engineer @voxmedia

prratek / serve.py

Created November 30, 2018 20:12

Modified serve file for Zeit deployment

	from starlette.applications import Starlette
	from starlette.responses import HTMLResponse, JSONResponse
	from starlette.staticfiles import StaticFiles
	from starlette.middleware.cors import CORSMiddleware
	import uvicorn, aiohttp, asyncio
	from io import BytesIO

	from fastai import *
	from fastai.vision import *

prratek / rf_tuned.py

Created September 26, 2018 18:35

	m = RandomForestRegressor(n_estimators=100, max_features='log2', n_jobs=-1)
	m.fit(X_train, y_train)
	print_score(m)

prratek / rf_tuning_max_features2.py

Created September 26, 2018 18:29

	m = RandomForestRegressor(max_features='sqrt', n_jobs=-1)
	m.fit(X_train, y_train)
	print_score(m)

prratek / rf_tuning_max_features.py

Created September 26, 2018 18:28

	m = RandomForestRegressor(max_features='log2', n_jobs=-1)
	m.fit(X_train, y_train)
	print_score(m)

prratek / rf_tuning_benchmark.py

Last active September 26, 2018 18:25

	m = RandomForestRegressor(n_jobs=-1)
	m.fit(X_train, y_train)
	print_score(m)

prratek / bagged_trees_r2.py

Created September 23, 2018 19:47

Increase in R^2 as we increase n_estimators

	import numpy as np
	import matplotlib.pyplot as plt
	from sklearn import metrics

	preds = np.stack([t.predict(X_valid) for t in m.estimators_])
	preds[:,0], np.mean(preds[:,0])
	plt.plot([metrics.r2_score(y_valid, np.mean(preds[:i+1], axis=0)) for i in range(10)]);

prratek / nycTaxi_rf_baseline.py

Last active September 23, 2018 00:18

Code for Random Forest Medium blog post

	import pandas as pd
	from sklearn.ensemble import RandomForestRegressor

	n = 100 # use every 100th row
	df = pd.read_csv('{PATH_TO_DATA}train.csv', skiprows=lambda i: i % n != 0)

	m = RandomForestRegressor() # instantiate the RandomForestRegressor objects
	m.fit(X_train, y_train) # train the model
	m.score(X_valid, y_valid) # score it on your validation set