Attend the Women in ML Symposium on December 7 Register now

monash_tsf

References:

weather

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/weather')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 3010
'train' 3010
'validation' 3010
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tourism_yearly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/tourism_yearly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 518
'train' 518
'validation' 518
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tourism_quarterly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/tourism_quarterly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 427
'train' 427
'validation' 427
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

tourism_monthly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/tourism_monthly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 366
'train' 366
'validation' 366
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

cif_2016

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/cif_2016')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 72
'train' 72
'validation' 72
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

london_smart_meters

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/london_smart_meters')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 5560
'train' 5560
'validation' 5560
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

australian_electricity_demand

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/australian_electricity_demand')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 5
'train' 5
'validation' 5
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

wind_farms_minutely

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/wind_farms_minutely')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 339
'train' 339
'validation' 339
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

bitcoin

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/bitcoin')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 18
'train' 18
'validation' 18
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

pedestrian_counts

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/pedestrian_counts')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 66
'train' 66
'validation' 66
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

vehicle_trips

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/vehicle_trips')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 329
'train' 329
'validation' 329
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

kdd_cup_2018

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/kdd_cup_2018')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 270
'train' 270
'validation' 270
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

nn5_daily

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/nn5_daily')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 111
'train' 111
'validation' 111
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

nn5_weekly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/nn5_weekly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 111
'train' 111
'validation' 111
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

kaggle_web_traffic

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/kaggle_web_traffic')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 145063
'train' 145063
'validation' 145063
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

kaggle_web_traffic_weekly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/kaggle_web_traffic_weekly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 145063
'train' 145063
'validation' 145063
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

solar_10_minutes

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/solar_10_minutes')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 137
'train' 137
'validation' 137
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

solar_weekly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/solar_weekly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 137
'train' 137
'validation' 137
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

car_parts

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/car_parts')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 2674
'train' 2674
'validation' 2674
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

fred_md

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/fred_md')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 107
'train' 107
'validation' 107
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

traffic_hourly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/traffic_hourly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 862
'train' 862
'validation' 862
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

traffic_weekly

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/traffic_weekly')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 862
'train' 862
'validation' 862
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

hospital

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/hospital')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 767
'train' 767
'validation' 767
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

covid_deaths

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/covid_deaths')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 266
'train' 266
'validation' 266
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

sunspot

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/sunspot')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 1
'train' 1
'validation' 1
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

saugeenday

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/saugeenday')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 1
'train' 1
'validation' 1
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

us_births

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/us_births')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 1
'train' 1
'validation' 1
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

solar_4_seconds

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/solar_4_seconds')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 1
'train' 1
'validation' 1
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

wind_4_seconds

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/wind_4_seconds')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 1
'train' 1
'validation' 1
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

rideshare

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/rideshare')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 156
'train' 156
'validation' 156
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

oikolab_weather

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/oikolab_weather')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 8
'train' 8
'validation' 8
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "dtype": "float32",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}

temperature_rain

Use the following command to load this dataset in TFDS:

ds = tfds.load('huggingface:monash_tsf/temperature_rain')
  • Description:
The first repository containing datasets of related time series for global forecasting.
Split Examples
'test' 422
'train' 422
'validation' 422
  • Features:
{
    "start": {
        "dtype": "timestamp[s]",
        "id": null,
        "_type": "Value"
    },
    "target": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_static_cat": {
        "feature": {
            "dtype": "uint64",
            "id": null,
            "_type": "Value"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "feat_dynamic_real": {
        "feature": {
            "feature": {
                "dtype": "float32",
                "id": null,
                "_type": "Value"
            },
            "length": -1,
            "id": null,
            "_type": "Sequence"
        },
        "length": -1,
        "id": null,
        "_type": "Sequence"
    },
    "item_id": {
        "dtype": "string",
        "id": null,
        "_type": "Value"
    }
}