search

sverhoeven / docker-cartodb

Dockerized CartoDB
BSD 3-Clause "New" or "Revised" License
148 stars 116 forks source link

Error in running Predict Trends and Volatility Analysis #56

Closed sindile closed 6 years ago

sindile commented 6 years ago

I am getting the following error "Analysis A1 failed: ValueError: numpy.dtype has the wrong size, try recompiling" when running the Predict Trends and Volatility Analysis I am using the following carto file - https://s3.amazonaws.com/cartodb/static/learn/predict-markov.zip from this link https://carto.com/learn/guides/analysis/predict-trends-and-volatility/

sverhoeven commented 6 years ago

Found cause during build a wheel of numpy==1.6.1 is installed after scikit-learn is compiled against numpy 1.13.3.

Build log:

pip install --upgrade ./crankshaft
Processing ./crankshaft
Collecting joblib==0.8.3 (from crankshaft==0.0.0)
  Downloading https://files.pythonhosted.org/packages/a4/84/26bb81ce7031e39c6566f58323bbe50ff3c80ce9b5180fdae29273ec2b9c/joblib-0.8.3-py2.py3-none-any.whl (81kB)
Collecting numpy==1.6.1 (from crankshaft==0.0.0)
  Downloading https://files.pythonhosted.org/packages/44/3b/63aa0464c495de6bac1f9f5a6c4626264ea07ad85b51cf6ebfdbbaa9fab4/numpy-1.6.1-cp27-cp27mu-manylinux1_x86_64.whl (13.6MB)
Collecting pysal==1.14.3 (from crankshaft==0.0.0)
  Downloading https://files.pythonhosted.org/packages/db/3f/2141de9a8849f7d24878c4e633291ed1f6d9f05f801ea8746576ecba1aed/PySAL-1.14.3.tar.gz (17.8MB)
Collecting scikit-learn==0.14.1 (from crankshaft==0.0.0)
  Downloading https://files.pythonhosted.org/packages/ee/a9/613e8107ea9ad60e03684a2eb2f824646f3d8f14e96ecfbf4077fc313e6c/scikit-learn-0.14.1.tar.gz (6.8MB)
Collecting scipy==0.14.0 (from crankshaft==0.0.0)
  Downloading https://files.pythonhosted.org/packages/13/b2/70a4992e0ef711122ba1d7fc0edc5fe121b105f768c020eff1ffac2ea8a0/scipy-0.14.0-cp27-cp27mu-manylinux1_x86_64.whl (35.3MB)
Building wheels for collected packages: crankshaft, pysal, scikit-learn
  Running setup.py bdist_wheel for crankshaft: started
  Running setup.py bdist_wheel for crankshaft: finished with status 'done'
  Stored in directory: /tmp/pip-ephem-wheel-cache-nmULvk/wheels/7e/de/83/422bf511b3bc8af205281bef65eacc4c74396d5f3ea0276cd3
  Running setup.py bdist_wheel for pysal: started
  Running setup.py bdist_wheel for pysal: finished with status 'done'
  Stored in directory: /root/.cache/pip/wheels/fa/dd/19/2665a24ef25064d2717f2e04aa173cd7876298ef4578bf9777
  Running setup.py bdist_wheel for scikit-learn: started
  Running setup.py bdist_wheel for scikit-learn: still running...
  Running setup.py bdist_wheel for scikit-learn: finished with status 'done'
  Stored in directory: /root/.cache/pip/wheels/43/e8/7d/c765a4b44b3752577aa2b16f5395d2e14477b7e0211c62cedc
Successfully built crankshaft pysal scikit-learn
Installing collected packages: joblib, numpy, scipy, pysal, scikit-learn, crankshaft
  Found existing installation: numpy 1.13.3
    Uninstalling numpy-1.13.3:
      Successfully uninstalled numpy-1.13.3
Successfully installed crankshaft-0.0.0 joblib-0.8.3 numpy-1.6.1 pysal-1.14.3 scikit-learn-0.14.1 scipy-0.14.0
sverhoeven commented 6 years ago

Analysis is now executing successfully.