12345678910111213141516171819202122232425262728293031323334353637383940414243444546474849505152535455565758596061626364656667686970717273747576777879808182838485868788899091929394 |
- import cudf
- import os
- import pandas as pd
- import urllib
- import tqdm
- from zipfile import ZipFile
- pbar = None
- def show_progress(block_num, block_size, total_size):
- global pbar
- if pbar is None:
- pbar = tqdm.tqdm(total=total_size / 1024, unit='kB')
- downloaded = block_num * block_size
- if downloaded < total_size:
- pbar.update(block_size / 1024)
- else:
- pbar.close()
- pbar = None
-
- def fetch_bike_dataset(years, data_dir="data"):
- """ Dowload bike dataset for a given year and return the list of files.
- """
- base_url = "https://s3.amazonaws.com/capitalbikeshare-data/"
- files = []
- for year in years:
- filename = str(year) + "-capitalbikeshare-tripdata.zip"
- filepath = os.path.join(data_dir, filename)
- if not os.path.isfile(filepath):
- urllib.request.urlretrieve(base_url+filename, filepath, reporthook=show_progress)
- with ZipFile(filepath) as myzip:
- files += [os.path.join(data_dir, name) for name in myzip.namelist()]
- myzip.extractall(data_dir)
-
- print("Files extracted: "+ str(files))
- return files
- def fetch_weather_dataset(data_dir='data'):
- base_url = 'https://archive.ics.uci.edu/ml/machine-learning-databases/00275/'
- fn = 'Bike-Sharing-Dataset.zip'
-
- if not os.path.isdir(data_dir):
- os.makedirs(data_dir)
- filepath = os.path.join(data_dir, fn)
-
- if not os.path.isfile(filepath):
- print(f'Downloading {base_url+fn} to {filepath}')
- urllib.request.urlretrieve(base_url+fn, filepath)
-
- files = []
- with ZipFile(filepath) as myzip:
- files = [os.path.join(data_dir, name) for name in myzip.namelist()]
- myzip.extractall(data_dir)
-
- # Extract weather features from the dataset
- # Note this weather dataset is already preprocessed.
- # We reverse the steps to provide a more interesting exercise.
- weather = cudf.read_csv(files[2], parse_dates=[1])
- out = cudf.DataFrame();
- out['Hour'] = weather['dteday'] + cudf.Series(pd.to_timedelta(weather['hr'].to_pandas(), unit='h'))
- out['Temperature'] = weather['temp'] * 47.0 -8
- out['Relative Temperature'] = weather['atemp'] * 66.0 - 16
- out['Rel. humidity'] = (weather['hum'] * 100).astype('int')
- out['Wind'] = weather['windspeed'] * 67
-
- # Spell out weather categories
- # - 1: Clear, Few clouds, Partly cloudy, Partly cloudy
- #- 2: Mist + Cloudy, Mist + Broken clouds, Mist + Few clouds, Mist
- #- 3: Light Snow, Light Rain + Thunderstorm + Scattered clouds, Light Rain + Scattered clouds
- # - 4: Heavy Rain + Ice Pallets + Thunderstorm + Mist, Snow + Fog
- out['Weather'] = 'Clear or Partly cloudy'
- out['Weather'][weather['weathersit']==2] = 'Mist or Cloudy'
- out['Weather'][weather['weathersit']==3] = 'Light Rain or Snow, Thunderstorm'
- out['Weather'][weather['weathersit']==4] = 'Heavy Rain, Snow + Fog, Ice'
-
- filepath = os.path.join(data_dir, 'weather2011-2012.csv')
- out.to_csv(filepath, index=False)
- print("Weather file saved at ", filepath)
- return filepath
- def read_bike_data_pandas(files):
- # Reads a list of files and concatenates them
- tables = []
- for filename in files:
- tmp_df = pd.read_csv(filename, usecols=[1], parse_dates=['Start date'])
- tables.append(tmp_df)
- merged_df = pd.concat(tables, ignore_index=True)
|