def init():
from sklearn.externals import joblib
global model, inputs_dc, prediction_dc
# The model we created in our linear_reg.py file is now a model.pkl
model = joblib.load('model.pkl')
inputs_dc = ModelDataCollector('model.pkl', identifier="inputs")
prediction_dc = ModelDataCollector('model.pkl', identifier="prediction")
def init():
from utils import *
global model, inputs_dc, prediction_dc
model = load_model('saved/model_json_week.h5')
word2vec = models.Word2Vec.load(saved_path + 'model_word2vec.pkl')
X, invalid_rows = texts_to_word_vec(word2vec, texts)
inputs_dc = ModelDataCollector('model.pkl',identifier="inputs")
prediction_dc = ModelDataCollector('model.pkl', identifier="prediction")
def init():
from sklearn.externals import joblib
import pickle
global model, inputs_dc, prediction_dc
model = joblib.load('testModel2.pkl')
inputs_dc = ModelDataCollector('testModel.pkl', identifier="inputs")
prediction_dc = ModelDataCollector('testModel.pkl',
identifier="prediction")
def init():
global inputs_dc, prediction_dc
from sklearn.externals import joblib
global model
model = joblib.load('model.pkl')
inputs_dc = ModelDataCollector("model.pkl", identifier="inputs")
prediction_dc = ModelDataCollector("model.pkl", identifier="prediction")
def init():
from sklearn.externals import joblib
global model
model = joblib.load('startupfunding.pkl')
global inputCollector, predictionCollector
inputCollector = ModelDataCollector('model.pkl', identifier="inputs")
predictionCollector = ModelDataCollector('model.pkl',
identifier="prediction")
def init():
global model, inputs_dc, prediction_dc
# Get the path to the model asset
# local_path = get_local_path('mymodel.model.link')
from sklearn.externals import joblib
inputs_dc = ModelDataCollector("model.pkl", identifier="inputs")
prediction_dc = ModelDataCollector("model.pkl", identifier="prediction")
# model = model_load_function(local_path)
model = joblib.load('model.pkl')
def init():
"""Web Service Initialization
Prepare the web service definition by authoring
init() and run() functions. Test the functions
before deploying the web service.
"""
import json
global inputs_dc, prediction_dc
global model, labels
model = load_cntk('model_cntk.pb')
labels = json.load(open('labels.json'))
inputs_dc = ModelDataCollector("model_tf.pb", identifier="inputs")
prediction_dc = ModelDataCollector("model_tf.pb", identifier="prediction")
def init():
global inputs_dc, prediction_dc
from sklearn.externals import joblib
# load the model from file into a global object
global model
with open('sgd_automated_learn_v2.pkl', 'rb') as fid:
model = pickle.load(fid, encoding='latin1')
input_t = np.array([[1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]])
print("model has been loaded ~~~~~~")
print(model.predict(input_t))
print("prediction test done")
inputs_dc = ModelDataCollector("sgd_automated_learn_v2.pkl",
identifier="inputs")
prediction_dc = ModelDataCollector("sgd_automated_learn_v2.pkl",
identifier="prediction")
# Download the model
if not os.path.exists(LOCAL_SYSTEM_DIRECTORY):
os.makedirs(LOCAL_SYSTEM_DIRECTORY)
if os.path.isfile(localFilePath):
os.remove(localFilePath)
az_blob_service = BlockBlobService(account_name=AZURE_STORAGE_ACCOUNT_NAME, account_key=AZURE_STORAGE_ACCOUNT_KEY)
az_blob_service.get_blob_to_path(AZURE_STORAGE_CONTAINER_NAME, AZURE_STORAGE_BLOB_NAME, localFilePath)
# load the model file
global model
model = joblib.load(localFilePath)
inputs_dc = ModelDataCollector(localFilePath, identifier="inputs")
prediction_dc = ModelDataCollector(localFilePath, identifier="prediction")
def upload_schema(localfile):
from azure.storage.blob import BlockBlobService
from azure.storage.blob import ContentSettings
az_blob_service = BlockBlobService(account_name=AZURE_STORAGE_ACCOUNT_NAME, account_key=AZURE_STORAGE_ACCOUNT_KEY)
az_blob_service.create_blob_from_path(
AZURE_STORAGE_CONTAINER_NAME,
AZURE_STORAGE_BLOB_NAME_SCHEMA,
localfile,
content_settings=ContentSettings(content_type='application/json'))
def run(input_df):
import json
