Getting Exception with datetime column for timeseries data with timegpt-1 on Azure

[prabirkrsarkar]

1. The Data :

print(df)

device_id,day,hour,datetime,message_count
Device4,2024-10-16,8,2024-10-16 08:00:00,7
Device4,2024-10-16,9,2024-10-16 09:00:00,8
Device4,2024-10-16,10,2024-10-16 10:00:00,9
Device4,2024-10-16,11,2024-10-16 11:00:00,7
Device4,2024-10-16,12,2024-10-16 12:00:00,9
Device4,2024-10-16,13,2024-10-16 13:00:00,13
Device4,2024-10-15,8,2024-10-15 08:00:00,4
Device4,2024-10-15,9,2024-10-15 09:00:00,9
Device4,2024-10-15,10,2024-10-15 10:00:00,6

2. Forecast :

forecast_df = nixtla_client.forecast(
df = df,
h = 24,
time_col="datetime",
target_col="message_count",
model="timegpt-1"
)

"name": "Exception",
"message": "Could not infer frequency of ds column. This could be due to inconsistent intervals. Please check your data for missing, duplicated or irregular timestamps",

The data (datetime & message_count) is that of footfall which is being aggregated on an hourly basis. Unable to understand why this is being flagged off as inconsistent interval and how to resolve this. Any help is appreciated. Thanks in advance.

Regards
Prabir Sarkar