datacanvasio / hypernets Goto Github PK

ERROR: Complete output from command 'D:\Anaconda\Anaconda3\python.exe' -u -c 'import setuptools, tokenize;file='"'"'C:\Users\John\AppData\Local\Temp\pip-install-4hgmcjg1\python-geohash\setup.py'"'"';f=getattr(tokenize, '"'"'open'"'"', open)(file);code=f.read().replace('"'"'\r\n'"'"', '"'"'\n'"'"');f.close();exec(compile(code, file, '"'"'exec'"'"'))' bdist_wheel -d 'C:\Users\John\AppData\Local\Temp\pip-wheel-gijlszf9' --python-tag cp37:
ERROR: running bdist_wheel
running build
running build_py
creating build
creating build\lib.win-amd64-3.7
copying geohash.py -> build\lib.win-amd64-3.7
copying quadtree.py -> build\lib.win-amd64-3.7
copying jpgrid.py -> build\lib.win-amd64-3.7
copying jpiarea.py -> build\lib.win-amd64-3.7
running build_ext
building '_geohash' extension
error: Microsoft Visual C++ 14.0 is required. Get it with "Microsoft Visual C++ Build Tools": https://visualstudio.microsoft.com/downloads/

ERROR: Failed building wheel for python-geohash
Running setup.py clean for python-geohash
Failed to build python-geohash
ERROR: spyder 3.3.6 requires pyqt5<5.13; python_version >= "3", which is not installed.
ERROR: spyder 3.3.6 requires pyqtwebengine<5.13; python_version >= "3", which is not installed.
ERROR: distributed 2021.11.1 has requirement dask==2021.11.1, but you'll have dask 2.1.0 which is incompatible.
ERROR: woodwork 0.9.0 has requirement pandas>=1.3.0, but you'll have pandas 1.2.3 which is incompatible.
ERROR: featuretools 1.2.0 has requirement dask[dataframe]>=2021.10.0, but you'll have dask 2.1.0 which is incompatible.
ERROR: featuretools 1.2.0 has requirement pandas<2.0.0,>=1.3.0, but you'll have pandas 1.2.3 which is incompatible.
Installing collected packages: psutil, pyyaml, tblib, distributed, woodwork, featuretools, python-geohash, hypernets
Found existing installation: psutil 5.6.3
Uninstalling psutil-5.6.3:
Successfully uninstalled psutil-5.6.3
Found existing installation: PyYAML 5.1.1
ERROR: Cannot uninstall 'PyYAML'. It is a distutils installed project and thus we cannot accurately determine which files belong to it which would lead to only a partial uninstall.

• research：

• HyperGBM

• Multi-task types, e.g. training, evaluation

• design

• implementation

• unit test

• integration of the dataset page, training page and experiment configuration page
• front-end: develop and test in the non-Notebook environment
• back-end: share the js program with Notebook
• unit test

an error about "from hypernets.frameworks.ml.datasets import dsutils", I think it should be "from deeptables.datasets import dsutils".

Please make sure that this is a feature request.

System information

• Hypernets version (you are using):
• Are you willing to contribute it (Yes/No):

Describe the feature and the current behavior/state.
I found that the user-defined evaluation metric function could not be set in the hypernets experiment, and calc_score() contained few metrics, such as no mape.

Will this change the current api? How?

Any Other info.

Thank you very much for your work! Would you please give some references for some of these strategies, for example,Hypernets/hypernets/searchers/evolution_searcher.py: mutate

The experiment report includes:

• dataset information
• feature transformation
• evaluation indicators
• confusion matrix
• resources monitoring
• importance of features
• prediction speed
• information of the ensemble models

The example of an exported experiment report:

• example

I noticed that using the pip install hyperkeras command did not download the hyperkeras related packages （i used python3.8）, so I directly downloaded and referenced the hyperkeras code library, and then used the code about defined in the CNN neural architecture search in the document you provided. However, I found the following error, which part of the definition is missing. I hope to receive your reply as soon as possible. Thank you

笔

On line 107 of voting's code, i.e.
if self.ensemble_size <= 0: size = predictions.shape[1]

Is it possible to replace it with self.ensemble_size <= 0 or predictions.shape[1] < self.ensemble_size: size = predictions.shape[1]?