Live loss plotting during training#

[1]:

if "google.colab" in str(get_ipython()):
    # uninstall preinstalled packages from Colab to avoid conflicts
    !pip uninstall -y torch notebook notebook_shim tensorflow tensorflow-datasets prophet torchaudio torchdata torchtext torchvision
    !pip install git+https://github.com/ourownstory/neural_prophet.git['live'] # may take a while
    #!pip install neuralprophet # much faster, but may not have the latest upgrades/bugfixes

import pandas as pd
from neuralprophet import NeuralProphet

[2]:

data_location = "https://raw.githubusercontent.com/ourownstory/neuralprophet-data/main/datasets/"
df = pd.read_csv(data_location + "retail_sales.csv")

[3]:

df_train, df_val = NeuralProphet().split_df(df, valid_p=0.2)

INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to [99.659]% of the data.
INFO - (NP.df_utils._infer_frequency) - Dataframe freq automatically defined as MS
INFO - (NP.df_utils.return_df_in_original_format) - Returning df with no ID column
INFO - (NP.df_utils.return_df_in_original_format) - Returning df with no ID column

[4]:

m = NeuralProphet()
metrics = m.fit(df_train, validation_df=df_val, progress="plot")

WARNING - (NP.forecaster.fit) - When Global modeling with local normalization, metrics are displayed in normalized scale.
INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to [99.574]% of the data.
INFO - (NP.df_utils._infer_frequency) - Dataframe freq automatically defined as MS
INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.
INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.
INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.
INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16
INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 338
WARNING - (NP.config.set_lr_finder_args) - Learning rate finder: The number of batches (15) is too small than the required number for the learning rate finder (213). The results might not be optimal.

../../_images/how-to-guides_feature-guides_Live_plot_during_training_5_341.png

[5]:

m = NeuralProphet()
metrics = m.fit(df_train, validation_df=df_val, progress="plot-all")

WARNING - (NP.forecaster.fit) - When Global modeling with local normalization, metrics are displayed in normalized scale.
INFO - (NP.df_utils._infer_frequency) - Major frequency MS corresponds to [99.574]% of the data.
INFO - (NP.df_utils._infer_frequency) - Dataframe freq automatically defined as MS
INFO - (NP.config.init_data_params) - Setting normalization to global as only one dataframe provided for training.
INFO - (NP.utils.set_auto_seasonalities) - Disabling weekly seasonality. Run NeuralProphet with weekly_seasonality=True to override this.
INFO - (NP.utils.set_auto_seasonalities) - Disabling daily seasonality. Run NeuralProphet with daily_seasonality=True to override this.
INFO - (NP.config.set_auto_batch_epoch) - Auto-set batch_size to 16
INFO - (NP.config.set_auto_batch_epoch) - Auto-set epochs to 338
WARNING - (NP.config.set_lr_finder_args) - Learning rate finder: The number of batches (15) is too small than the required number for the learning rate finder (213). The results might not be optimal.