Change column type change value from another column (Python script)

Question

Hello.&nbsp;I made a python script and it runs in power query, it returns one date and one float column. When I try to change column types according to their values, they change. Changing the column type also changes the values in the same column. Here an example.&nbsp;This is what returns python script.&nbsp;&nbsp;&nbsp;I try to change the type of date column to date type and&nbsp;and the values in the prediction column change.&nbsp;In report view the values are also different from those shown in power query.&nbsp;In the python script, a model is created with neural networks from the keras library and returns the predictions of the initial data.&nbsp;I believe is due to keras, how can I avoid this?&nbsp;I appreciate any help. Thank you for your ideas and time.&nbsp;&nbsp;&nbsp;

edhans · Answer

what happens if in the step before you do a the change type if you wrap it in a Table.Buffer() function? Just put Table.Buffer() around the entire line, and&nbsp;then change the type to Date.

inm · Answer

Hi.I tried what you said but that also change the values.

v-yingjl · Answer

Hi&nbsp;inm&nbsp;,
Could you please consider sharing a sample file without any sesentive information or the sample python script query code for further discussion?
&nbsp;
Best Regards,Community Support Team _ Yingjie Li

inm · Answer

Python script:import pandas as pdimport numpy as npfrom keras.layers import Dense, LSTMfrom keras.models import Sequentialfrom keras.preprocessing.sequence import TimeseriesGenerator&nbsp;def get_prediction_dates(num_prediction, df):&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;last_date = df['date'].max()&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;date_test = pd.date_range(last_date, periods=num_prediction+1, freq='M').to_frame(index=False, &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;name='date')&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;date_test = date_test['date'].apply(lambda x : x.replace(day=1))&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;return date_test&nbsp;def predict(num_prediction, model):&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;prediction_list = arr[-look_back:]&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;for _ in range(num_prediction):&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;x = prediction_list[-look_back:]&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;x = x.reshape((1, look_back, 1))&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;out = model.predict(x)[0][0]&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;prediction_list = np.append(prediction_list, out)&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp;prediction_list = prediction_list[look_back-1:]return prediction_list&nbsp;look_back = 12batch_size = 4arr = dataset['quantity'].valuesarr = arr.reshape((-1, 1))generator = TimeseriesGenerator(arr, arr, length=look_back, batch_size = batch_size)model = Sequential()model.add(LSTM(128,activation='linear',input_shape=(look_back, 1),name="LSTM"))model.add(Dense(1, activation='linear', name="Dense"))model.compile(optimizer='adam', loss='mse', metrics=['mean_absolute_percentage_error'])model.fit(generator, epochs=70)num_prediction = 6forecast = predict(num_prediction, model)forecast_dates = get_prediction_dates(num_prediction, dataset)dataset = pd.DataFrame(list(zip(forecast_dates, forecast)), columns=['date', 'prediction'])dataset['date'] = pd.to_datetime(dataset['date'])&nbsp;I placed a seed, but is there any other way to prevent this from happening?

Forum Discussion

Change column type change value from another column (Python script)

4 Replies

Related content

Publish .pbip using Python script

Power BI Python ADO.NET: Python Script Error

Python Script in Power Query Function

Python script error

python script error