Dataset Preview
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code: DatasetGenerationError
Exception: ArrowInvalid
Message: Schema at index 1 was different:
a: string
b: string
n_series: int64
n_configs: int64
identical_configs: int64
differing: struct<kuramoto/K1: string, kuramoto/K1.5: string, mackey_glass/tau17: string, mackey_glass/tau20: string, mackey_glass/tau30: string, tails/fbm/H0.8: string>
series_identical: int64
max_abs_diff_over_context_std: struct<kuramoto/K1: double, kuramoto/K1.5: double, mackey_glass/tau17: double, mackey_glass/tau20: double, mackey_glass/tau30: double, tails/fbm/H0.8: double>
vs
model: string
set: string
split: string
overall: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>
by_family: struct<chaos: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, in_prior: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, regime: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, tails: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>>
by_failure_type: struct<bursty: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, chaos_short_reach: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_in_prior: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_old_switch: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_unknowable: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_unpredictable: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, coupled_oscillators: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, delay_oscillator: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, heavy_tail_walk: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, long_memory: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, multifractal: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, periodic_map: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, recent_switch: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, smooth_chaos: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, switching: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>>
provenance: struct<script: string, git: string, time_utc: timestamp[s], host: string, torch: string, threads: int64, cuda: bool, argv: list<item: string>>
Traceback: Traceback (most recent call last):
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1876, in _prepare_split_single
num_examples, num_bytes = writer.finalize()
~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 830, in finalize
self.write_rows_on_file()
~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 689, in write_rows_on_file
table = pa.concat_tables(self.current_rows)
File "pyarrow/table.pxi", line 6268, in pyarrow.lib.concat_tables
File "pyarrow/error.pxi", line 155, in pyarrow.lib.pyarrow_internal_check_status
File "pyarrow/error.pxi", line 92, in pyarrow.lib.check_status
raise convert_status(status)
pyarrow.lib.ArrowInvalid: Schema at index 1 was different:
a: string
b: string
n_series: int64
n_configs: int64
identical_configs: int64
differing: struct<kuramoto/K1: string, kuramoto/K1.5: string, mackey_glass/tau17: string, mackey_glass/tau20: string, mackey_glass/tau30: string, tails/fbm/H0.8: string>
series_identical: int64
max_abs_diff_over_context_std: struct<kuramoto/K1: double, kuramoto/K1.5: double, mackey_glass/tau17: double, mackey_glass/tau20: double, mackey_glass/tau30: double, tails/fbm/H0.8: double>
vs
model: string
set: string
split: string
overall: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>
by_family: struct<chaos: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, in_prior: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, regime: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, tails: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>>
by_failure_type: struct<bursty: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, chaos_short_reach: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_in_prior: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_old_switch: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_unknowable: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, control_unpredictable: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, coupled_oscillators: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, delay_oscillator: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, heavy_tail_walk: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, long_memory: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, multifractal: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, periodic_map: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, recent_switch: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, smooth_chaos: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>, switching: struct<n: int64, mase_rel_snaive: double, scrps_rel_snaive: double, mase_rel_oracle: double, cov80: double>>
provenance: struct<script: string, git: string, time_utc: timestamp[s], host: string, torch: string, threads: int64, cuda: bool, argv: list<item: string>>
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1379, in compute_config_parquet_and_info_response
parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
~~~~~~~~~~~~~~~~~~~~~~~~~^
builder, max_dataset_size_bytes=max_dataset_size_bytes
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
)
^
File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 956, in stream_convert_to_parquet
builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1702, in _prepare_split
for job_id, done, content in self._prepare_split_single(
~~~~~~~~~~~~~~~~~~~~~~~~~~^
gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
):
^
File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1887, in _prepare_split_single
raise DatasetGenerationError("An error occurred while generating the dataset") from e
datasets.exceptions.DatasetGenerationError: An error occurred while generating the datasetNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
set string | split string | grid string | config_id string | seed_index int64 | T int64 | H int64 | failure_type string | reason string |
|---|---|---|---|---|---|---|---|---|
diag | public | otp.grids.chaos | mackey_glass/tau17 | 0 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 1 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 2 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 3 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 4 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 5 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 6 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 7 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 8 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 9 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 10 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau17 | 11 | 2,048 | 128 | delay_oscillator | smooth delay-feedback oscillation; Timestral leans toward a periodic template and a nearest-neighbour forecaster beats it |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 0 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 1 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 2 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 3 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 4 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 5 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 6 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 7 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 8 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 9 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 10 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau20 | 11 | 2,048 | 128 | delay_oscillator | longer delay, less regular cycles; same pattern as tau = 17, larger gap |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 0 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 1 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 2 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 3 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 4 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 5 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 6 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 7 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 8 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 9 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 10 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | mackey_glass/tau30 | 11 | 2,048 | 128 | delay_oscillator | long-delay chaos in the delay equation; the largest Mackey-Glass gap in the map |
diag | public | otp.grids.chaos | rossler/c5.7 | 0 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 1 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 2 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 3 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 4 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 5 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 6 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 7 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 8 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 9 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 10 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c5.7 | 11 | 2,048 | 128 | smooth_chaos | Rossler chaos: near-periodic loops whose timing drifts |
diag | public | otp.grids.chaos | rossler/c9 | 0 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 1 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 2 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 3 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 4 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 5 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 6 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 7 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 8 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 9 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 10 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | rossler/c9 | 11 | 2,048 | 128 | smooth_chaos | Rossler with more irregular loop sizes |
diag | public | otp.grids.chaos | kuramoto/K0 | 0 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 1 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 2 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 3 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 4 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 5 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 6 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 7 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 8 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 9 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 10 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K0 | 11 | 2,048 | 128 | coupled_oscillators | incoherent coupled oscillators: a sum of many drifting phases |
diag | public | otp.grids.chaos | kuramoto/K1 | 0 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 1 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 2 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 3 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 4 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 5 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 6 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 7 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 8 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 9 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 10 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1 | 11 | 2,048 | 128 | coupled_oscillators | partially synchronized oscillators; the largest gap to the oracle in the map |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 0 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 1 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 2 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 3 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 4 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 5 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 6 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 7 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 8 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 9 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 10 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | kuramoto/K1.5 | 11 | 2,048 | 128 | coupled_oscillators | near the synchronization threshold |
diag | public | otp.grids.chaos | lorenz63/rho28 | 0 | 2,048 | 128 | chaos_short_reach | Lorenz chaos: the gap lives within one to three Lyapunov times |
diag | public | otp.grids.chaos | lorenz63/rho28 | 1 | 2,048 | 128 | chaos_short_reach | Lorenz chaos: the gap lives within one to three Lyapunov times |
diag | public | otp.grids.chaos | lorenz63/rho28 | 2 | 2,048 | 128 | chaos_short_reach | Lorenz chaos: the gap lives within one to three Lyapunov times |
diag | public | otp.grids.chaos | lorenz63/rho28 | 3 | 2,048 | 128 | chaos_short_reach | Lorenz chaos: the gap lives within one to three Lyapunov times |
End of preview.
Off-the-Prior benchmarks
Synthetic forecasting benchmarks built to find where time-series foundation models trained on simple synthetic priors stop working: chaotic and delay-feedback dynamics, coupled oscillators, regime switches, heavy tails, long memory, bursts, learned mixtures of these, and in-prior controls. Every series includes the oracle's 21 quantiles: a Monte Carlo continuation of the true simulator from its true state, so models can be compared with the best achievable forecast, not only with each other.
- diag: 372 series (372 public, 0 held-out), each tagged with why it is hard.
- lb: 3776 series (1888 public, 1888 held-out), all simulator configs, several context lengths and horizons.
Files are NumPy .npz archives (load with numpy.load; the Hugging Face datasets loader is not used).
Each .npz file holds series_id, context (N, T), future (N, H), oracle_q (N, H, 21), period (N,) and a
JSON meta string per series (generator, parameters, tags, reason). bench/build.py rebuilds the files from the
config lists in configs/; *_SHA256SUMS lists content hashes of the published files. The published files were
built on Windows 11, Python 3.12, numpy 2.4.6, scipy 1.18.0; a rebuild on Linux, Python 3.13, numpy 2.5.3, scipy 1.18.1 matched array-for-array for 25 of 31 diagnostic
configs, and in 6 (kuramoto/K1, kuramoto/K1.5, mackey_glass/tau17, mackey_glass/tau20, mackey_glass/tau30, tails/fbm/H0.8) platform floating-point differences change the simulated paths
(data/cross_platform_diag.json). Use these files, not a rebuild, for reported scores.
Evaluation: python bench/eval.py --set diag --predictions <dir>; see the repository README.
The seeds and generator code are public. The held-out leaderboard files exist for convenience, not secrecy.
License: the data are generated by the project's own code (Apache 2.0). No third-party data are included.
- Downloads last month
- 79