Date issues related to monthly prediction

[shubhashish1]

Discussed in https://github.com/awslabs/gluon-ts/discussions/1577

^{Originally posted by **shubhashish1** June 21, 2021} Hi, I am doing a monthly predictions for a list of items and the dates of each month available is the first date of the month i. e 1-1-2017 . But while training the same in the first date of every month I am getting the following error: **GluonTSDateBoundsError: Out of bounds nanosecond timestamp: 9043-01-31 00:00:00** 

[lostella]

Hi @shubhashish1 it's very hard to understand from the screenshot where the error may be coming from. Could you please provide a running example, possibly with fake data, to reproduce the issue?

[StatMixedML]

@shubhashish1 It seems like that your frequency is misspecified, or that your date format is ambiguous and not aligned with the gluonts supported frequencies.

Also note that, currently, gluonts only supports MonthEnd-type of frequencies, see here. If you want to change this to allow a MonthStart-type format (e.g., 2017-01-01) you would need to change it in the time_feature sub-folder.

[mliuzzi-amorelie]

Hi, I am having exactly the same problem: when training on data from 2017 to 2021 I get the error: gluonts.core.exception.GluonTSDateBoundsError: Out of bounds nanosecond timestamp: 3230-03-31 00:00:00. When I predict for daily frequency, everything runs fine, but since I changed it to Monthly I am getting this error. Any help is really appreciated. Please let me know if I need to specify more info. Thanks!

[mbohlkeschneider]

@mliuzzi-amorelie Can you provide a code snipped that produces the error? Thank you!

[mliuzzi-amorelie]

@mbohlkeschneider df_train["timestamp"].max() '2020-12-30T00:00:00.000Z'

df_train["timestamp"].min() '2020-01-01T00:00:00.000Z'

And this is the function that produces the error:

import sagemaker
from sagemaker.mxnet import MXNet

mxnet_estimator = MXNet(entry_point='blog_train_algos.py',
                        role=sagemaker.get_execution_role(),
                        instance_type='ml.m5.large',
                        instance_count=1,
                        framework_version='1.7.0', 
                        py_version='py3',
                        hyperparameters={'bucket': bucket,
                            'seq': trial.trial_name,
                            'algo': "DeepAR",             
                            'freq': "M", 
                            'prediction_length': 6, 
                            'epochs': 10,
                            'learning_rate': 1e-3,
                            'hybridize': False,
                            'num_batches_per_epoch': 10,
                         })

[mliuzzi-amorelie]

And the entrypoint script 'blog_train_algos.py' is this:

import os
os.system('pip install pandas')
os.system('pip install gluonts')
import pandas as pd
import pathlib
import gluonts
import numpy as np
import argparse
import json
import boto3
from mxnet import gpu, cpu
from mxnet.context import num_gpus
from gluonts.dataset.util import to_pandas
from gluonts.model.deepar import DeepAREstimator
from gluonts.model.simple_feedforward import SimpleFeedForwardEstimator
from gluonts.model.lstnet import LSTNetEstimator
from gluonts.model.seq2seq import MQCNNEstimator
from gluonts.model.transformer import TransformerEstimator
from gluonts.evaluation.backtest import make_evaluation_predictions, backtest_metrics
from gluonts.evaluation import Evaluator
from gluonts.model.predictor import Predictor
from gluonts.dataset.common import ListDataset
from gluonts.mx.trainer import Trainer
from gluonts.dataset.multivariate_grouper import MultivariateGrouper
from smdebug.mxnet import Hook

s3 = boto3.client("s3")

def uploadDirectory(model_dir,prefix,bucket):
    for root,dirs,files in os.walk(model_dir):
        for file in files:
            print(os.path.join(root,file))
            print(prefix+file)
            s3.upload_file(os.path.join(root,file),bucket,prefix+file)

def train(bucket, seq, algo, freq, prediction_length, epochs, learning_rate, hybridize, num_batches_per_epoch):

    #create train dataset
    df = pd.read_csv(filepath_or_buffer=os.environ['SM_CHANNEL_TRAIN'] + "/train.csv", header=0, index_col=0)

    training_data = ListDataset([{"start": df.index[0], 
                                  "target": df.demand[:],
                                 "item_id": df.item_id[:]}], 
                                  freq=freq)

    #create test dataset
    df = pd.read_csv(filepath_or_buffer=os.environ['SM_CHANNEL_TEST'] + "/test.csv", header=0, index_col=0)

    test_data = ListDataset([{"start": df.index[0], 
                              "target": df.demand[:],
                              "item_id": 'ALN-33812-01'}], 
                              freq=freq)

    hook = Hook.create_from_json_file()
    #determine estimators##################################
    if algo == "DeepAR":
        estimator = DeepAREstimator(freq=freq,  
                                       prediction_length=prediction_length,
                                       context_length=1,
                                       trainer=Trainer(ctx="cpu",
                                            epochs=epochs,
                                            learning_rate=learning_rate,
                                            hybridize=hybridize,
                                            num_batches_per_epoch=num_batches_per_epoch
                                         ))

        #train the model
        predictor = estimator.train(training_data=training_data)
        print("DeepAR training is complete SUCCESS")
    elif algo == "SFeedFwd":
        estimator = SimpleFeedForwardEstimator(freq=freq,  
                                       prediction_length=prediction_length,
                                       trainer=Trainer(ctx="cpu",
                                            epochs=epochs,
                                            learning_rate=learning_rate,
                                            hybridize=hybridize,
                                            num_batches_per_epoch=num_batches_per_epoch
                                         ))

        #train the model
        predictor = estimator.train(training_data=training_data)
        print("training is complete SUCCESS")
    elif algo == "lstnet":
        # Needed for LSTNet ONLY
        grouper = MultivariateGrouper(max_target_dim=6)
        training_data = grouper(training_data)
        test_data = grouper(test_data)
        context_length = prediction_length
        num_series = 1
        skip_size = 1
        ar_window = 1
        channels = 4

        estimator = LSTNetEstimator(freq=freq,  
                                       prediction_length=prediction_length,
                                       context_length=context_length,
                                       num_series=num_series,
                                       skip_size=skip_size,
                                       ar_window=ar_window,
                                       channels=channels, 
                                       trainer=Trainer(ctx="cpu",
                                            epochs=epochs,
                                            learning_rate=learning_rate,
                                            hybridize=hybridize,
                                            num_batches_per_epoch=num_batches_per_epoch
                                         ))

        #train the model
        predictor = estimator.train(training_data=training_data)
        print("training is complete SUCCESS")
    elif algo == "seq2seq":
        estimator = MQCNNEstimator(freq=freq,  
                                       prediction_length=prediction_length,
                                       trainer=Trainer(ctx="cpu",
                                            epochs=epochs,
                                            learning_rate=learning_rate,
                                            hybridize=hybridize,
                                            num_batches_per_epoch=num_batches_per_epoch
                                         ))

        #train the model
        predictor = estimator.train(training_data=training_data)
        print("training is complete SUCCESS")
    else:
        estimator = TransformerEstimator(freq=freq,  
                                       prediction_length=prediction_length,
                                       trainer=Trainer(ctx="cpu",
                                            epochs=epochs,
                                            learning_rate=learning_rate,
                                            hybridize=hybridize,
                                            num_batches_per_epoch=num_batches_per_epoch
                                         ))

        #train the model
        predictor = estimator.train(training_data=training_data)
        print("training is complete SUCCESS")

    ###################################################

    #evaluate trained model on test data
    forecast_it, ts_it = make_evaluation_predictions(test_data, predictor,  num_samples=100)
    print("EVALUATION is complete SUCCESS")
    forecasts = list(forecast_it)
    tss = list(ts_it)
    evaluator = Evaluator(quantiles=[0.1, 0.5, 0.9])
    agg_metrics, item_metrics = evaluator(iter(tss), iter(forecasts), num_series=len(test_data))
    print("METRICS retrieved SUCCESS")
    #bucket = "bwp-sandbox"

    mainpref = "gluonts/blog-models/"
    prefix = mainpref + str(seq) + "/"
    agg_df = pd.DataFrame(agg_metrics, index=[0])
    file = "metrics"+str(seq)+".csv"
    os.system('mkdir metrics')
    cspath = os.path.join('metrics', file)
    agg_df.to_csv(cspath)
    s3.upload_file(cspath,bucket,mainpref+"metrics/"+file)

    hook.save_scalar("MAPE", agg_metrics["MAPE"], sm_metric=True)
    hook.save_scalar("RMSE", agg_metrics["RMSE"], sm_metric=True)
    hook.save_scalar("MASE", agg_metrics["MASE"], sm_metric=True)
    hook.save_scalar("MSE", agg_metrics["MSE"], sm_metric=True)

    print("MAPE:", agg_metrics["MAPE"])

    #save the model
    predictor.serialize(pathlib.Path(os.environ['SM_MODEL_DIR'])) 

    uploadDirectory(os.environ['SM_MODEL_DIR'], prefix, bucket)

    return predictor

def model_fn(model_dir):
    path = pathlib.Path(model_dir)   
    predictor = Predictor.deserialize(path)

    return predictor

def transform_fn(model, data, content_type, output_content_type):

    data = json.loads(data)
    df = pd.DataFrame(data)

    test_data = ListDataset([{"start": df.index[0], 
                              "target": df.demand[:]}], 
                               freq=freq)

    forecast_it, ts_it = make_evaluation_predictions(test_data, model,  num_samples=100)  
    #agg_metrics, item_metrics = Evaluator()(ts_it, forecast_it, num_series=len(test_data))
    #response_body = json.dumps(agg_metrics)
    response_body = json.dumps({'predictions':list(forecast_it)[0].samples.tolist()[0]})
    return response_body, output_content_type

def parse_args():
    parser = argparse.ArgumentParser()
    parser.add_argument('--bucket', type=str, default="")
    parser.add_argument('--seq', type=str, default="sample")
    parser.add_argument('--algo', type=str, default="DeepAR")
    parser.add_argument('--freq', type=str, default='M')
    parser.add_argument('--prediction_length', type=int, default=6)
    parser.add_argument('--epochs', type=int, default=10)
    parser.add_argument('--learning_rate', type=float, default=1e-3)
    parser.add_argument('--hybridize', type=bool, default=True)
    parser.add_argument('--num_batches_per_epoch', type=int, default=10)    

    return parser.parse_args()

if __name__ == '__main__':
    args = parse_args()
    train(args.bucket, args.seq, args.algo, args.freq, args.prediction_length, args.epochs, args.learning_rate, args.hybridize, args.num_batches_per_epoch)

[mliuzzi-amorelie]

@shubhashish1 did you manage to solve this problem for your data?