def train(cls, handler, app_id, krequest, dfe):
autorun = make_autorun(dfe, feature_type_estimation=False)
descriptions = autorun.execute()
model = autorun.result()
score = 0 if model is None else model.score
messages = []
for i, d in enumerate(descriptions):
if i == len(descriptions) - 1 and model is None:
message = {"error": 1, "message": d.desc}
else:
message = {"error": 0, "message": d.desc}
messages.append(message)
image = b"" if model is None else model.describe().picture.to_base64()
result = {
"score": score,
"messages": messages,
"image": image.decode("utf-8")
}
predictor = autorun.to_predictor()
predictor.save_to_env(krequest.env, app_id)
handler.write(result)
def test_autorun(self):
print("AutoRun --------------")
app = Application()
dfe = app.load(self.APP_ID)
dfe.target = "家賃"
dfe.drop("物件名")
dfe.drop("担当者")
autorun = make_autorun(dfe, feature_type_estimation=False)
descriptions = autorun.execute()
for d in descriptions:
print(d)
def test_analyst(self):
df = pd.read_csv(self.FILE_NAME)
autorun = make_autorun(df)
autorun.dfe.target = "Survived"
autorun.dfe.drop("PassengerId")
autorun.dfe.drop("Ticket")
autorun.dfe.drop("Name")
descriptions = autorun.execute()
for d in descriptions:
print(d)
result = autorun.result()
print(result.describe().picture.path)
def test_autorun_from_request(self):
print("AutoRun from Request --------------")
request = json.loads(KINTONE_REQUEST.replace("###", str(self.APP_ID)))
krequest = kintoneRequest()
dfe = krequest.request_to_dfe(request)
dfe.to_categorical(["向き"])
autorun = make_autorun(dfe, feature_type_estimation=False)
descriptions = autorun.execute()
for d in descriptions:
print(d)
print("Predict from kintone Request --------------")
record = json.loads(KINTONE_RECORD.replace("###", str(self.APP_ID)))
df = krequest.record_to_df(record)
predictor = autorun.to_predictor()
pred = predictor.predict(df)
print("estimated value: {}".format(pred))
def test_classifier_predictor(self):
df = pd.read_csv(self.FILE_NAME)
autorun = make_autorun(df)
autorun.dfe.target = "Survived"
autorun.dfe.drop("PassengerId")
autorun.dfe.drop("Ticket")
autorun.dfe.drop("Name")
descriptions = autorun.execute()
for d in descriptions:
print(d)
predictor = autorun.to_predictor()
path = os.path.dirname(__file__)
saved_path = predictor.save(path, "test_predictor")
predictor = Predictor.load(path, "test_predictor")
shutil.rmtree(saved_path)
df2 = pd.read_csv(self.FILE_NAME)
y_true = df2["Survived"]
pred = predictor.predict(df2)
score = accuracy_score(y_true, pred)
print(score)