jens-andersson-2-wcar

jens-andersson-2-wcar

positions_df_byhex = df_subset.groupby("hex_id", as_index=False).agg({
    "longitude": "mean",
    "latitude": "mean",
    "vector_component_x": "mean",
    "vector_component_y": "mean",
    "bearing": "count", # "mean" would be wrong here; we need circmean
    "speed_kmh": "mean"
  })

jens-andersson-2-wcar

arrow_tip_x = size_of_arrow_x * math.cos(bearing_rad) + center_lat
arrow_tip_y = size_of_arrow_y * math.sin(bearing_rad) + center_long

arrow_back_left_x = size_of_arrow_x * math.cos(bearing_rad + 2*math.pi * 135.0 / 360.0 ) + center_lat
arrow_back_left_y = size_of_arrow_y * math.sin(bearing_rad + 2*math.pi * 135.0 / 360.0 ) + center_long

arrow_back_x = size_of_arrow_x * 0.2 * math.cos(bearing_rad + 2*math.pi * 180.0 / 360.0 ) + center_lat
arrow_back_y = size_of_arrow_y * 0.2 * math.sin(bearing_rad + 2*math.pi * 180.0 / 360.0 ) + center_long

arrow_back_right_x = size_of_arrow_x * math.cos(bearing_rad + 2*math.pi * 225.0 / 360.0 ) + center_lat

jens-andersson-2-wcar

import numpy as np
import scipy.stats

# compass bearings in degrees:
data = [2, 356]

# average bearing:
print(np.mean(data)) # == 179.0, aka "wrong"
print(scipy.stats.circmean(data, high=360, low=0)) # == 359.0, aka "correct"

jens-andersson-2-wcar

jens-andersson-2-wcar

!conda install -y folium geojson

import folium
from geojson import Feature, Point, FeatureCollection
import json

def hexagons_dataframe_to_geojson(df_hex, file_output = None, column_name = "value"):
    """
    Produce the GeoJSON for a dataframe, constructing the geometry from the "hex_id" column
    and with a property matching the one in column_name

jens-andersson-2-wcar

dataset2001byhexid = dataset2001.groupby("hex_id", as_index=False).agg({"pop_2001": "sum"})

jens-andersson-2-wcar

!conda install -y h3-py

import h3

hex_ids = dataset2001.apply(lambda row: h3.geo_to_h3(row.lat, row.long, resolution), axis = 1)
dataset2001 = dataset2001.assign(hex_id=hex_ids.values)

jens-andersson-2-wcar

from sagemaker.pipeline import PipelineModel
from sagemaker.model import Model
from sagemaker.transformer import Transformer

import sagemaker

# The location of the input dataset, WITHOUT the column to predict (and WITHOUT headers)
batch_input = 's3://YOURBUCKET/YOURFOLDER/predictmeplease.csv'

# The location to store the results of the batch transform job (will be stored under this folder, with a single value per line)