/auto_ml

auto_ml based on hyper opt-sklearn

Primary LanguagePython

AutoMachineLearning

  • Three API for u currently

Upload_dataset

  • files = {'file': ('files', strings, "application/octet-stream")}
  • strings is the binary data with dict-form, which own given keys and values
  • strings = pickle.dumps({'X_train':X_train, 'y_train':y_train,'X_test':X_test, 'y_test':y_test},2)
  • Return data_id or Error Info

Sample:

  • requests.post(url, files=files)

Check_dataset

  • dataid is the code get from upload_dataset API
  • Parameter Form:
  • None
  • Return dict

Sample:

  • requests.get(url + dataid)

Auto_ml

  • dataid is the code get from upload_dataset API
  • Parameter Form:
  • params = {'regressor': 'Null','preprocessing': 'Null', 'max_evals': 5,'trial_timeout': 10,'seed': 'Null'}
  • Return Model parameter

Sample:

  • r = requests.post(url, params=params)