在 Python 中运行 OSMNX 时，如何解决 Intel MK 致命错误？

Question

I am running Python 3.6.1 with Anaconda 4.4.0 (64-bit) on Windows 10. I am trying to run OSMNX sample code ( seen here )

My code starts by successfully importing packages:

import requests
import matplotlib.cm as cm
import matplotlib.colors as colors
import pandas as pd
import numpy as np
from datetime import datetime
from dateutil.parser import parse
import geopandas
import osmnx 

When I run the following code I find no error:

osmnx.save_gdf_shapefile(city)
city = osmnx.project_gdf(city)
fig, ax = osmnx.plot_shape(city, figsize=(3,3))

I find an error when I run the following:

G = osmnx.graph_from_point((37.79, -122.41), distance=1, network_type='all')

This is the error I get:

Projected the GeoDataFrame "geometry to project" to UTM-10 in 0.01 seconds
Projected the GeoDataFrame "geometry to project" to default_crs in 0.01 seconds
Created bounding box 1 meters in each direction from (37.79, -122.41): 37.79000901274339,37.78999098725659,-122.4099886430464,-122.41001135695363
Projected the GeoDataFrame "geometry to project" to UTM-10 in 0.01 seconds
Projected the GeoDataFrame "geometry to project" to default_crs in 0.01 seconds
Projected the GeoDataFrame "geometry to project" to UTM-10 in 0.01 seconds
Projected the GeoDataFrame "geometry to project" to default_crs in 0.01 seconds
Requesting network data within bounding box from API in 1 request(s)
Retrieved response from cache file "cache\ab03dfa5a6c968bcbd2e10dd360bd8f6.json" for URL "http://overpass-api.de/api/interpreter?data=%5Bout%3Ajson%5D%5Btimeout%3A180%5D%3B%28way%5B%22highway%22%5D%5B%22area%22%21~%22yes%22%5D%5B%22highway%22%21~%22proposed%7Cconstruction%7Cabandoned%7Cplatform%7Craceway%22%5D%5B%22service%22%21~%22private%22%5D%5B%22access%22%21~%22private%22%5D%2837.785485%2C-122.415690%2C37.794515%2C-122.404310%29%3B%3E%3B%29%3Bout%3B"
Got all network data within bounding box from API in 1 request(s) and 0.01 seconds
Creating networkx graph from downloaded OSM data...
Graph was not connected, retained only the largest weakly connected component (1,496 of 1,511 total nodes) in 0.06 seconds
Created graph with 1,496 nodes and 3,149 edges in 0.10 seconds
Intel MKL FATAL ERROR: Cannot load mkl_intel_thread.dll. 

How do I solve this error? Thanks in advance!

Answer 1

How did you install OSMnx? If you install it in a conda env per its install instructions you should not experience this sort of missing dll/system path issue. Your best bet to solve this would be to install it according to that documentation.