Spaces:
Sleeping
Sleeping
zhouziyu02 Cursor commited on
Commit ·
1d5b407
1
Parent(s): 8fcb36a
Refresh live eval results and show raw metric aggregation.
Browse filesRe-run all 6 TSFM models on 10 TS-Bench tasks with fresh data; remove baseline normalization from leaderboard display.
Co-authored-by: Cursor <cursoragent@cursor.com>
- app.py +4 -6
- results/chronos_2/all_results.csv +10 -8
- results/chronos_2/online_meta.json +29 -17
- results/chronos_bolt_base/all_results.csv +10 -8
- results/chronos_bolt_base/online_meta.json +29 -17
- results/chronos_bolt_mini/all_results.csv +10 -8
- results/chronos_bolt_mini/online_meta.json +29 -17
- results/chronos_bolt_tiny/all_results.csv +10 -8
- results/chronos_bolt_tiny/online_meta.json +29 -17
- results/dataset_properties.csv +3 -1
- results/moirai_1_1_small/all_results.csv +10 -8
- results/moirai_1_1_small/online_meta.json +29 -17
- results/online_status.json +176 -104
- results/timesfm_2_5/all_results.csv +10 -8
- results/timesfm_2_5/online_meta.json +29 -17
- src/about.py +15 -8
- src/utils.py +3 -27
app.py
CHANGED
|
@@ -18,7 +18,6 @@ from src.display.css_html_js import (
|
|
| 18 |
)
|
| 19 |
from src.utils import (
|
| 20 |
get_grouped_dfs,
|
| 21 |
-
load_baseline_name,
|
| 22 |
prepare_ranks_table,
|
| 23 |
prepare_values_table,
|
| 24 |
)
|
|
@@ -34,14 +33,14 @@ def load_online_status(results_path: str = RESULTS_PATH) -> dict:
|
|
| 34 |
return json.loads(status_path.read_text())
|
| 35 |
|
| 36 |
|
| 37 |
-
def build_status_line(status: dict
|
| 38 |
finished = status.get("finished_at", "n/a")
|
| 39 |
data_source = status.get("data_source", "n/a")
|
| 40 |
ts_bench_root = status.get("ts_bench_root", "")
|
| 41 |
root_hint = f" · TS-Bench: `{ts_bench_root}`" if ts_bench_root else ""
|
| 42 |
return (
|
| 43 |
f"**Live benchmark** · Data: `{data_source}`{root_hint} · "
|
| 44 |
-
f"
|
| 45 |
f"Auto-refresh every {REFRESH_SECONDS}s"
|
| 46 |
)
|
| 47 |
|
|
@@ -49,12 +48,11 @@ def build_status_line(status: dict, baseline: str) -> str:
|
|
| 49 |
def refresh_leaderboard() -> tuple:
|
| 50 |
grouped = get_grouped_dfs(RESULTS_PATH)
|
| 51 |
status = load_online_status(RESULTS_PATH)
|
| 52 |
-
baseline = load_baseline_name(RESULTS_PATH)
|
| 53 |
|
| 54 |
outputs: list = [
|
| 55 |
prepare_values_table(grouped["overall_values"]),
|
| 56 |
prepare_ranks_table(grouped["overall_ranks"]),
|
| 57 |
-
build_status_line(status
|
| 58 |
]
|
| 59 |
for domain in DOMAINS:
|
| 60 |
values_df = grouped["domain_values"].get(domain)
|
|
@@ -91,7 +89,7 @@ with demo:
|
|
| 91 |
value=INITIAL_OUTPUTS[0],
|
| 92 |
interactive=False,
|
| 93 |
wrap=True,
|
| 94 |
-
label="Metric values (
|
| 95 |
)
|
| 96 |
overall_ranks_df = gr.Dataframe(
|
| 97 |
value=INITIAL_OUTPUTS[1],
|
|
|
|
| 18 |
)
|
| 19 |
from src.utils import (
|
| 20 |
get_grouped_dfs,
|
|
|
|
| 21 |
prepare_ranks_table,
|
| 22 |
prepare_values_table,
|
| 23 |
)
|
|
|
|
| 33 |
return json.loads(status_path.read_text())
|
| 34 |
|
| 35 |
|
| 36 |
+
def build_status_line(status: dict) -> str:
|
| 37 |
finished = status.get("finished_at", "n/a")
|
| 38 |
data_source = status.get("data_source", "n/a")
|
| 39 |
ts_bench_root = status.get("ts_bench_root", "")
|
| 40 |
root_hint = f" · TS-Bench: `{ts_bench_root}`" if ts_bench_root else ""
|
| 41 |
return (
|
| 42 |
f"**Live benchmark** · Data: `{data_source}`{root_hint} · "
|
| 43 |
+
f"Last eval: `{finished}` · "
|
| 44 |
f"Auto-refresh every {REFRESH_SECONDS}s"
|
| 45 |
)
|
| 46 |
|
|
|
|
| 48 |
def refresh_leaderboard() -> tuple:
|
| 49 |
grouped = get_grouped_dfs(RESULTS_PATH)
|
| 50 |
status = load_online_status(RESULTS_PATH)
|
|
|
|
| 51 |
|
| 52 |
outputs: list = [
|
| 53 |
prepare_values_table(grouped["overall_values"]),
|
| 54 |
prepare_ranks_table(grouped["overall_ranks"]),
|
| 55 |
+
build_status_line(status),
|
| 56 |
]
|
| 57 |
for domain in DOMAINS:
|
| 58 |
values_df = grouped["domain_values"].get(domain)
|
|
|
|
| 89 |
value=INITIAL_OUTPUTS[0],
|
| 90 |
interactive=False,
|
| 91 |
wrap=True,
|
| 92 |
+
label="Metric values (geometric mean across datasets, lower is better)",
|
| 93 |
)
|
| 94 |
overall_ranks_df = gr.Dataframe(
|
| 95 |
value=INITIAL_OUTPUTS[1],
|
results/chronos_2/all_results.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
-
binance_btcusdt_1h_close/1h/short,Chronos-2,
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
+
binance_btcusdt_1h_close/1h/short,Chronos-2,4380093.155741374,4380093.155741374,1850.0364583333333,2.3838636016637054,0.025076568408860828,0.024678861943730448,66.88197607794382,2092.8672092947927,0.0282057298197708,0.024933081405608456,0.021564816825717217,Finance,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd/1h/short,Chronos-2,3302102.223780314,3302102.223780314,1589.05078125,1.9819221083493386,0.02156161218573524,0.021260322353283848,46.8612864887045,1817.1687383895623,0.02450082462177794,0.021425117923286136,0.017372480048838304,Finance,1
|
| 4 |
+
nasa_power_shanghai_hourly_T2M/1h/short,Chronos-2,16.264854419926603,16.264854419926603,3.1164464155832925,0.9942700533878908,0.1112300523205632,0.12196958213989656,21.94742601097775,4.032970917317233,0.15713129036201698,0.12142196327823566,0.1048847735590834,Weather,1
|
| 5 |
+
noaa_coops_sf_water_level_water_level/6min/short,Chronos-2,0.06003561507700905,0.06003561507700905,0.213998693972826,8.070613535689246,1.013472949369841,0.597409693051528,96.8635045441028,0.24502166246478913,0.9537009413978706,0.832949845522854,0.5562847402156419,Ocean,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height/10min/short,Chronos-2,0.009589138295815758,0.009589138295815758,0.07559254765510559,0.5440371647347783,0.0288622829034601,0.029130192091777325,4.1629569029245,0.09792414562208729,0.0375428043484963,0.028981169136490237,0.03618823474141707,Ocean,1
|
| 7 |
+
nws_ksfo_observations_temperature/1h/short,Chronos-2,0.003463902878100574,0.003463902878100574,0.02116549015045166,0.4010301182731177,0.0016484734142401373,0.0016379871537986845,9.278132686642117,0.0588549307883424,0.00453311396422428,0.0016302046009655346,0.0018722179194657846,Weather,1
|
| 8 |
+
open_meteo_shanghai_air_quality_hourly_pm2_5/1h/short,Chronos-2,121.62235993895926,121.62235993895926,9.762574990590414,0.5251049426581348,0.25052453302473104,0.21224658419640266,4.581651077587976,11.028252805361294,0.2247033380106543,0.1989148441445373,0.2278692478910483,Air Quality,1
|
| 9 |
+
open_meteo_shanghai_hourly_temperature_2m/1h/short,Chronos-2,0.30751845326813054,0.30751845326813054,0.4332726796468099,0.23597305034316118,0.017424389083777924,0.017404320116509994,2.091343357081799,0.5545434638223865,0.0221301007506678,0.0172905618380289,0.022378411607540734,Weather,1
|
| 10 |
+
usgs_potomac_iv_usgs_00060/15min/short,Chronos-2,5269.190091451009,5269.190091451009,65.57242838541667,nan,0.0025808082601376345,0.0025768785670678665,nan,72.58918715243344,0.0028545641351112607,0.002578630642716697,0.001904930492972491,Hydrology,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews/1D/short,Chronos-2,4063.1116605430143,4063.1116605430143,63.214759826660156,1.3546019962855749,0.2031273434135618,0.22642697598080944,5.089582116263253,63.74254199938228,0.20528999033617482,0.20359020878151418,0.18650881592714202,Web,1
|
results/chronos_2/online_meta.json
CHANGED
|
@@ -1,55 +1,67 @@
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-2",
|
| 3 |
"api_model_id": "amazon/chronos-2",
|
| 4 |
-
"evaluated_at": "2026-05-
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
-
"dataset": "
|
| 8 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"context_length": 96,
|
| 10 |
"prediction_length": 24
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 14 |
-
"data_fetched_at": "2026-05-
|
| 15 |
-
"context_length":
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"dataset": "
|
| 20 |
-
"data_fetched_at": "2026-05-
|
| 21 |
"context_length": 96,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"dataset": "
|
| 26 |
-
"data_fetched_at": "2026-05-
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"dataset": "
|
| 32 |
-
"data_fetched_at": "2026-05-
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"dataset": "
|
| 38 |
-
"data_fetched_at": "2026-05-
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"dataset": "
|
| 44 |
-
"data_fetched_at": "2026-05-
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"dataset": "
|
| 50 |
-
"data_fetched_at": "2026-05-
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
]
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-2",
|
| 3 |
"api_model_id": "amazon/chronos-2",
|
| 4 |
+
"evaluated_at": "2026-05-28T12:20:18.799459+00:00",
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 8 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 9 |
+
"context_length": 96,
|
| 10 |
+
"prediction_length": 24
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 14 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 15 |
"context_length": 96,
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 20 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 21 |
+
"context_length": 83,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 26 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 32 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 38 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 44 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 50 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 56 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 57 |
"context_length": 96,
|
| 58 |
"prediction_length": 24
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 62 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 63 |
+
"context_length": 4,
|
| 64 |
+
"prediction_length": 2
|
| 65 |
}
|
| 66 |
]
|
| 67 |
}
|
results/chronos_bolt_base/all_results.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
-
binance_btcusdt_1h_close/1h/short,Chronos-Bolt-Base,
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
+
binance_btcusdt_1h_close/1h/short,Chronos-Bolt-Base,4627089.7309748335,4627089.7309748335,1874.4404296875,2.4153093273872144,0.025415879545412644,0.024995872396728593,52.461912288031094,2151.067114474775,0.028990094347893916,0.025261975575046464,0.019576547766102168,Finance,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd/1h/short,Chronos-Bolt-Base,4946482.715515137,4946482.715515137,1928.9296875,2.4058314801622274,0.026177410818833124,0.02572777590076968,48.45609391950561,2224.0689547572792,0.02998704647293313,0.02600769371756978,0.019592513629850913,Finance,1
|
| 4 |
+
nasa_power_shanghai_hourly_T2M/1h/short,Chronos-Bolt-Base,11.688720341001195,11.688720341001195,2.729196389516195,0.8707219307033777,0.09833320637961653,0.10621373079641616,8.073066633299149,3.4188770584800494,0.13320516681172243,0.10633405475219919,0.08541267410550227,Weather,1
|
| 5 |
+
noaa_coops_sf_water_level_water_level/6min/short,Chronos-Bolt-Base,0.03234704147219288,0.03234704147219288,0.13382656003038088,5.047051581319789,0.6706533574633848,0.41720830815031723,22.83294145173187,0.1798528328167029,0.7000434747071089,0.5208948262007407,0.4173842468196957,Ocean,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height/10min/short,Chronos-Bolt-Base,0.04078452699226176,0.04078452699226176,0.17054141561190286,1.2273811519446596,0.06768574491893825,0.06480219526685438,7.064554223996088,0.20195179373370706,0.07742560970848361,0.06538329192416381,0.061584868474800714,Ocean,1
|
| 7 |
+
nws_ksfo_observations_temperature/1h/short,Chronos-Bolt-Base,0.0037005670213829944,0.0037005670213829944,0.01999191443125407,0.3787939590286682,0.0015589197272527502,0.0015477403017569464,13.350006060847901,0.06083228601148402,0.004685413464949044,0.0015398136615912097,0.0017142914417336535,Weather,1
|
| 8 |
+
open_meteo_shanghai_air_quality_hourly_pm2_5/1h/short,Chronos-Bolt-Base,38.0265807971964,38.0265807971964,4.774677435557048,0.25681817793213635,0.1013508384060875,0.09818999823413017,3.9438365445644235,6.166569613423366,0.12564535839599894,0.0972852161289089,0.17859769229816153,Air Quality,1
|
| 9 |
+
open_meteo_shanghai_hourly_temperature_2m/1h/short,Chronos-Bolt-Base,0.6043374313377777,0.6043374313377777,0.6502586205800375,0.3541499785661307,0.026396943053871125,0.026084632371007877,3.108420925181368,0.7773914273631899,0.03102326820637271,0.02594979424739109,0.027723991507018,Weather,1
|
| 10 |
+
usgs_potomac_iv_usgs_00060/15min/short,Chronos-Bolt-Base,4639.297880808513,4639.297880808513,62.649576822916664,nan,0.0024655696006461968,0.0024621813211091674,nan,68.11239153640483,0.002678514495942513,0.002463689732508602,0.0015585257620341544,Hydrology,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews/1D/short,Chronos-Bolt-Base,4028.8496973256115,4028.8496973256115,63.05271911621094,1.3511296953473773,0.2026653811521117,0.22577298837325582,4.94127665928432,63.473220316331926,0.20442260971443454,0.20306833853852155,0.181499555051636,Web,1
|
results/chronos_bolt_base/online_meta.json
CHANGED
|
@@ -1,55 +1,67 @@
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-Bolt-Base",
|
| 3 |
"api_model_id": "amazon/chronos-bolt-base",
|
| 4 |
-
"evaluated_at": "2026-05-
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
-
"dataset": "
|
| 8 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"context_length": 96,
|
| 10 |
"prediction_length": 24
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 14 |
-
"data_fetched_at": "2026-05-
|
| 15 |
-
"context_length":
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"dataset": "
|
| 20 |
-
"data_fetched_at": "2026-05-
|
| 21 |
"context_length": 96,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"dataset": "
|
| 26 |
-
"data_fetched_at": "2026-05-
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"dataset": "
|
| 32 |
-
"data_fetched_at": "2026-05-
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"dataset": "
|
| 38 |
-
"data_fetched_at": "2026-05-
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"dataset": "
|
| 44 |
-
"data_fetched_at": "2026-05-
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"dataset": "
|
| 50 |
-
"data_fetched_at": "2026-05-
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
]
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-Bolt-Base",
|
| 3 |
"api_model_id": "amazon/chronos-bolt-base",
|
| 4 |
+
"evaluated_at": "2026-05-28T12:19:28.056546+00:00",
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 8 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 9 |
+
"context_length": 96,
|
| 10 |
+
"prediction_length": 24
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 14 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 15 |
"context_length": 96,
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 20 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 21 |
+
"context_length": 83,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 26 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 32 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 38 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 44 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 50 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 56 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 57 |
"context_length": 96,
|
| 58 |
"prediction_length": 24
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 62 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 63 |
+
"context_length": 4,
|
| 64 |
+
"prediction_length": 2
|
| 65 |
}
|
| 66 |
]
|
| 67 |
}
|
results/chronos_bolt_mini/all_results.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
-
binance_btcusdt_1h_close/1h/short,Chronos-Bolt-Mini,
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
+
binance_btcusdt_1h_close/1h/short,Chronos-Bolt-Mini,3679822.6813456216,3679822.6813456216,1654.4537760416667,2.1318456290822945,0.022439195282325037,0.02210438779696162,36.894830953375646,1918.2863918991923,0.02585289092712348,0.022297198789814756,0.016141385649750175,Finance,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd/1h/short,Chronos-Bolt-Mini,3435441.7622146606,3435441.7622146606,1580.2054036458333,1.9708898306919778,0.02145509434972552,0.021141595171611006,22.12416891501253,1853.4944732085555,0.024990603275380642,0.02130585599630342,0.013890711488634161,Finance,1
|
| 4 |
+
nasa_power_shanghai_hourly_T2M/1h/short,Chronos-Bolt-Mini,9.13818853845593,9.13818853845593,2.457893212636312,0.784165453167993,0.08935921081957772,0.09565064369787313,7.269570607721318,3.0229436876091373,0.1177789406529396,0.09576362934213734,0.07988338981433957,Weather,1
|
| 5 |
+
noaa_coops_sf_water_level_water_level/6min/short,Chronos-Bolt-Mini,0.02307176451407955,0.02307176451407955,0.1410230112572511,5.318454063280177,0.6473759859576461,0.45994482734131564,19.641841542644524,0.15189392520466233,0.5912186620687938,0.5489056650822877,0.38075484208964183,Ocean,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height/10min/short,Chronos-Bolt-Mini,0.03395013763280493,0.03395013763280493,0.15618878602981567,1.1240857326666478,0.06198749864085912,0.05956084295321493,7.4643506648433195,0.18425563121056826,0.0706411382882342,0.059880686199460895,0.0619203976492256,Ocean,1
|
| 7 |
+
nws_ksfo_observations_temperature/1h/short,Chronos-Bolt-Mini,0.003293288668335966,0.003293288668335966,0.019572933514912922,0.3708553776284693,0.0015254847226304179,0.0015155839450093345,12.664317868508965,0.05738718209091614,0.004420065286277438,0.0015075429883074389,0.0017640636910896848,Weather,1
|
| 8 |
+
open_meteo_shanghai_air_quality_hourly_pm2_5/1h/short,Chronos-Bolt-Mini,58.57930447065731,58.57930447065731,6.592350959777832,0.35458637460433845,0.15445332516570973,0.14256432500281468,4.4046256992472514,7.65371181000809,0.1559462430027276,0.13432075707220698,0.20333559743785112,Air Quality,1
|
| 9 |
+
open_meteo_shanghai_hourly_temperature_2m/1h/short,Chronos-Bolt-Mini,0.8828099140087033,0.8828099140087033,0.7188208103179932,0.39149096453335225,0.028733217601040043,0.028834013771296452,2.773892978739956,0.9395796475066408,0.03749569442091312,0.02868589748468996,0.02969891826111223,Weather,1
|
| 10 |
+
usgs_potomac_iv_usgs_00060/15min/short,Chronos-Bolt-Mini,4843.679802894592,4843.679802894592,64.029296875,nan,0.002519860464793693,0.0025163275141648713,nan,69.59655022265538,0.002736878920766392,0.0025179471161723742,0.001616091514077776,Hydrology,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews/1D/short,Chronos-Bolt-Mini,4577.245442453423,4577.245442453423,67.47705841064453,1.445936965942383,0.21708162745704251,0.24360223675984644,8.007688195364816,67.65534304438506,0.21789160400768137,0.21731741839176982,0.23123530744333173,Web,1
|
results/chronos_bolt_mini/online_meta.json
CHANGED
|
@@ -1,55 +1,67 @@
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-Bolt-Mini",
|
| 3 |
"api_model_id": "amazon/chronos-bolt-mini",
|
| 4 |
-
"evaluated_at": "2026-05-
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
-
"dataset": "
|
| 8 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"context_length": 96,
|
| 10 |
"prediction_length": 24
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 14 |
-
"data_fetched_at": "2026-05-
|
| 15 |
-
"context_length":
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"dataset": "
|
| 20 |
-
"data_fetched_at": "2026-05-
|
| 21 |
"context_length": 96,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"dataset": "
|
| 26 |
-
"data_fetched_at": "2026-05-
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"dataset": "
|
| 32 |
-
"data_fetched_at": "2026-05-
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"dataset": "
|
| 38 |
-
"data_fetched_at": "2026-05-
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"dataset": "
|
| 44 |
-
"data_fetched_at": "2026-05-
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"dataset": "
|
| 50 |
-
"data_fetched_at": "2026-05-
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
]
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-Bolt-Mini",
|
| 3 |
"api_model_id": "amazon/chronos-bolt-mini",
|
| 4 |
+
"evaluated_at": "2026-05-28T12:18:55.358936+00:00",
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 8 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 9 |
+
"context_length": 96,
|
| 10 |
+
"prediction_length": 24
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 14 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 15 |
"context_length": 96,
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 20 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 21 |
+
"context_length": 83,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 26 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 32 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 38 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 44 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 50 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 56 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 57 |
"context_length": 96,
|
| 58 |
"prediction_length": 24
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 62 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 63 |
+
"context_length": 4,
|
| 64 |
+
"prediction_length": 2
|
| 65 |
}
|
| 66 |
]
|
| 67 |
}
|
results/chronos_bolt_tiny/all_results.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
-
binance_btcusdt_1h_close/1h/short,Chronos-Bolt-Tiny,
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
+
binance_btcusdt_1h_close/1h/short,Chronos-Bolt-Tiny,4495006.537943522,4495006.537943522,1834.72265625,2.3641310092463494,0.024881456748435934,0.02447329243393806,42.788615799160766,2120.143046575755,0.028573328343718278,0.024726696135602966,0.018015038439380838,Finance,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd/1h/short,Chronos-Bolt-Tiny,4406332.5283203125,4406332.5283203125,1807.1061197916667,2.2538886819785158,0.024529248640723117,0.02412824693875963,37.43977008949355,2099.1266108361147,0.028302453076857404,0.02436515067565812,0.017333996733501764,Finance,1
|
| 4 |
+
nasa_power_shanghai_hourly_T2M/1h/short,Chronos-Bolt-Tiny,7.912413253015529,7.912413253015529,2.2673874696095786,0.7233865627167552,0.08212307663246772,0.0874627393188528,2.2266680874402787,2.812901216362837,0.10959533476675763,0.088341207054233,0.07361846408709673,Weather,1
|
| 5 |
+
noaa_coops_sf_water_level_water_level/6min/short,Chronos-Bolt-Tiny,0.0028608058548383144,0.0028608058548383144,0.049041210363308586,1.8495096807224467,0.225792045612945,0.19658526579294686,13.967965661021097,0.05348650161338199,0.20818619230490099,0.19088372848461752,0.20820748121803842,Ocean,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height/10min/short,Chronos-Bolt-Tiny,0.04937945652725754,0.04937945652725754,0.18734135230382284,1.348289762768264,0.07458714345199068,0.07090175042748963,7.0800555112695145,0.22221488817641705,0.0851942084062309,0.07182416237838214,0.06322130203989622,Ocean,1
|
| 7 |
+
nws_ksfo_observations_temperature/1h/short,Chronos-Bolt-Tiny,0.003276212487397364,0.003276212487397364,0.02068618933359782,0.39194863412581116,0.001611035489054791,0.0016012503189502235,12.192836604810884,0.057238208282556886,0.004408591052225484,0.0015932879790811604,0.001700157311234707,Weather,1
|
| 8 |
+
open_meteo_shanghai_air_quality_hourly_pm2_5/1h/short,Chronos-Bolt-Tiny,63.19570465734099,63.19570465734099,6.434090296427409,0.34607392203585513,0.14174756363352167,0.1365677615769388,4.458499596558531,7.949572608470281,0.1619744788596924,0.13109615749526107,0.20445340630382589,Air Quality,1
|
| 9 |
+
open_meteo_shanghai_hourly_temperature_2m/1h/short,Chronos-Bolt-Tiny,0.5948811111434225,0.5948811111434225,0.6484154860178629,0.35314615324951076,0.026197955231929627,0.025978455964565026,2.995656525072318,0.7712853629775573,0.030779594213512538,0.025876240493323176,0.02777387188201828,Weather,1
|
| 10 |
+
usgs_potomac_iv_usgs_00060/15min/short,Chronos-Bolt-Tiny,3149.2314387957254,3149.2314387957254,54.302978515625,nan,0.0021364303572376296,0.0021344827808550478,nan,56.11801349652111,0.002206836513053427,0.002135460403694904,0.0014279900766016713,Hydrology,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews/1D/short,Chronos-Bolt-Tiny,4641.08992965892,4641.08992965892,67.62164306640625,1.4490352085658482,0.21732069855015004,0.24415806200991483,8.605713272094727,68.125545353112,0.21940594316622222,0.21778306945702497,0.2325466885637525,Web,1
|
results/chronos_bolt_tiny/online_meta.json
CHANGED
|
@@ -1,55 +1,67 @@
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-Bolt-Tiny",
|
| 3 |
"api_model_id": "amazon/chronos-bolt-tiny",
|
| 4 |
-
"evaluated_at": "2026-05-
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
-
"dataset": "
|
| 8 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"context_length": 96,
|
| 10 |
"prediction_length": 24
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 14 |
-
"data_fetched_at": "2026-05-
|
| 15 |
-
"context_length":
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"dataset": "
|
| 20 |
-
"data_fetched_at": "2026-05-
|
| 21 |
"context_length": 96,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"dataset": "
|
| 26 |
-
"data_fetched_at": "2026-05-
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"dataset": "
|
| 32 |
-
"data_fetched_at": "2026-05-
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"dataset": "
|
| 38 |
-
"data_fetched_at": "2026-05-
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"dataset": "
|
| 44 |
-
"data_fetched_at": "2026-05-
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"dataset": "
|
| 50 |
-
"data_fetched_at": "2026-05-
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
]
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model": "Chronos-Bolt-Tiny",
|
| 3 |
"api_model_id": "amazon/chronos-bolt-tiny",
|
| 4 |
+
"evaluated_at": "2026-05-28T12:18:05.053451+00:00",
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 8 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 9 |
+
"context_length": 96,
|
| 10 |
+
"prediction_length": 24
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 14 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 15 |
"context_length": 96,
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 20 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 21 |
+
"context_length": 83,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 26 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 32 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 38 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 44 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 50 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 56 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 57 |
"context_length": 96,
|
| 58 |
"prediction_length": 24
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 62 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 63 |
+
"context_length": 4,
|
| 64 |
+
"prediction_length": 2
|
| 65 |
}
|
| 66 |
]
|
| 67 |
}
|
results/dataset_properties.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,domain,frequency,num_variates
|
| 2 |
binance_btcusdt_1h_close,Finance,1h,1
|
|
|
|
| 3 |
nasa_power_shanghai_hourly_t2m,Weather,1h,1
|
| 4 |
noaa_coops_sf_water_level_water_level,Ocean,6min,1
|
|
|
|
| 5 |
nws_ksfo_observations_temperature,Weather,1h,1
|
| 6 |
open_meteo_shanghai_air_quality_hourly_pm2_5,Air Quality,1h,1
|
| 7 |
open_meteo_shanghai_hourly_temperature_2m,Weather,1h,1
|
| 8 |
-
usgs_earthquake_all_week_hourly_earthquake_count,Hazards,1h,1
|
| 9 |
usgs_potomac_iv_usgs_00060,Hydrology,15min,1
|
|
|
|
|
|
| 1 |
dataset,domain,frequency,num_variates
|
| 2 |
binance_btcusdt_1h_close,Finance,1h,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd,Finance,1h,1
|
| 4 |
nasa_power_shanghai_hourly_t2m,Weather,1h,1
|
| 5 |
noaa_coops_sf_water_level_water_level,Ocean,6min,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height,Ocean,10min,1
|
| 7 |
nws_ksfo_observations_temperature,Weather,1h,1
|
| 8 |
open_meteo_shanghai_air_quality_hourly_pm2_5,Air Quality,1h,1
|
| 9 |
open_meteo_shanghai_hourly_temperature_2m,Weather,1h,1
|
|
|
|
| 10 |
usgs_potomac_iv_usgs_00060,Hydrology,15min,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews,Web,1D,1
|
results/moirai_1_1_small/all_results.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
-
binance_btcusdt_1h_close/1h/short,Moirai-1.1-Small,
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
+
binance_btcusdt_1h_close/1h/short,Moirai-1.1-Small,8869335.108332315,7410361.406759898,2589.4026692708335,3.3365735823857965,0.035022303348886334,0.0343546582322679,33.181524624215,2978.1428958886972,0.04013665726762046,0.034897575803988286,0.022626002293161866,Finance,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd/1h/short,Moirai-1.1-Small,3513908.5660069785,3177372.7269617715,1577.9192708333333,1.9680384824423314,0.021403528918522274,0.021113633378368957,50.807128803692336,1874.5422283872344,0.02527438944637319,0.021275032144936267,0.0176526858138831,Finance,1
|
| 4 |
+
nasa_power_shanghai_hourly_T2M/1h/short,Moirai-1.1-Small,16.46809493020707,17.762606350991785,3.2289379835128784,1.03015932672555,0.1150119196766957,0.12686051090764325,15.835757291519093,4.058090059400736,0.15810997413863923,0.12580482285892922,0.10702413857726764,Weather,1
|
| 5 |
+
noaa_coops_sf_water_level_water_level/6min/short,Moirai-1.1-Small,0.005106911324939271,0.005023066128869232,0.061788178980350494,2.3302409204783667,0.26746783672360525,0.22650807762100097,22.07225150197471,0.071462656856146,0.2781550105933736,0.24049891698571166,0.309963690852684,Ocean,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height/10min/short,Moirai-1.1-Small,0.011788767184261436,0.009929538068242039,0.07437371214230855,0.5352652442580991,0.029046001335686772,0.02862860060252284,5.27525164074997,0.10857608937635135,0.04162661674999375,0.028513883944475068,0.039676113202017554,Ocean,1
|
| 7 |
+
nws_ksfo_observations_temperature/1h/short,Moirai-1.1-Small,0.009181221596918476,0.0033276888748332567,0.017116785049438477,0.3243178534509565,0.0013366897324967708,0.001326615176037771,15.141958610055587,0.09581869127116314,0.007380130120228421,0.0013183659600123755,0.002086598079706136,Weather,1
|
| 8 |
+
open_meteo_shanghai_air_quality_hourly_pm2_5/1h/short,Moirai-1.1-Small,187.9900532822021,186.3770823168015,9.616911133130392,0.5172700413546742,0.1997350052588335,0.24096720588809792,6.676616536919641,13.710946476527509,0.2793638752154284,0.1959469075568973,0.21197237663449722,Air Quality,1
|
| 9 |
+
open_meteo_shanghai_hourly_temperature_2m/1h/short,Moirai-1.1-Small,0.7451148641512191,0.7336263607690702,0.7014484405517578,0.38202946077832767,0.028982091534334033,0.02845294724981401,3.3017538202957617,0.8632003615332996,0.03444763524408459,0.027992620368853683,0.026208127846003952,Weather,1
|
| 10 |
+
usgs_potomac_iv_usgs_00060/15min/short,Moirai-1.1-Small,10798.155609925589,6040.659053166707,68.469970703125,nan,0.0026951572531489633,0.0026905473764017203,nan,103.91417424935632,0.004086416814655992,0.002692576268843192,0.0026380449775952625,Hydrology,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews/1D/short,Moirai-1.1-Small,3089.0964928199537,2815.892555135797,52.98386764526367,1.1353685923985073,0.17051746000903956,0.18643314794028967,2.6726661137172156,55.579640992183045,0.17900045408110482,0.17064047550809555,0.13484585527060686,Web,1
|
results/moirai_1_1_small/online_meta.json
CHANGED
|
@@ -1,55 +1,67 @@
|
|
| 1 |
{
|
| 2 |
"model": "Moirai-1.1-Small",
|
| 3 |
"api_model_id": "Salesforce/moirai-1.1-R-small",
|
| 4 |
-
"evaluated_at": "2026-05-
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
-
"dataset": "
|
| 8 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"context_length": 96,
|
| 10 |
"prediction_length": 24
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 14 |
-
"data_fetched_at": "2026-05-
|
| 15 |
-
"context_length":
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"dataset": "
|
| 20 |
-
"data_fetched_at": "2026-05-
|
| 21 |
"context_length": 96,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"dataset": "
|
| 26 |
-
"data_fetched_at": "2026-05-
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"dataset": "
|
| 32 |
-
"data_fetched_at": "2026-05-
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"dataset": "
|
| 38 |
-
"data_fetched_at": "2026-05-
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"dataset": "
|
| 44 |
-
"data_fetched_at": "2026-05-
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"dataset": "
|
| 50 |
-
"data_fetched_at": "2026-05-
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
]
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model": "Moirai-1.1-Small",
|
| 3 |
"api_model_id": "Salesforce/moirai-1.1-R-small",
|
| 4 |
+
"evaluated_at": "2026-05-28T12:21:49.771336+00:00",
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 8 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 9 |
+
"context_length": 96,
|
| 10 |
+
"prediction_length": 24
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 14 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 15 |
"context_length": 96,
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 20 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 21 |
+
"context_length": 83,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 26 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 32 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 38 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 44 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 50 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 56 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 57 |
"context_length": 96,
|
| 58 |
"prediction_length": 24
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 62 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 63 |
+
"context_length": 4,
|
| 64 |
+
"prediction_length": 2
|
| 65 |
}
|
| 66 |
]
|
| 67 |
}
|
results/online_status.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
{
|
| 2 |
"status": "ok",
|
| 3 |
-
"started_at": "2026-05-
|
| 4 |
-
"finished_at": "2026-05-
|
| 5 |
"data_source": "ts_bench",
|
| 6 |
"data_config": "configs/datasets/ts_bench.yaml",
|
| 7 |
"ts_bench_root": "/Users/zhouziyu/Desktop/TS-Bench",
|
|
@@ -10,330 +10,402 @@
|
|
| 10 |
"Chronos-Bolt-Tiny": {
|
| 11 |
"model": "Chronos-Bolt-Tiny",
|
| 12 |
"api_model_id": "amazon/chronos-bolt-tiny",
|
| 13 |
-
"evaluated_at": "2026-05-
|
| 14 |
"datasets": [
|
| 15 |
{
|
| 16 |
-
"dataset": "
|
| 17 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
"context_length": 96,
|
| 19 |
"prediction_length": 24
|
| 20 |
},
|
| 21 |
{
|
| 22 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 23 |
-
"data_fetched_at": "2026-05-
|
| 24 |
-
"context_length":
|
| 25 |
"prediction_length": 24
|
| 26 |
},
|
| 27 |
{
|
| 28 |
-
"dataset": "
|
| 29 |
-
"data_fetched_at": "2026-05-
|
| 30 |
"context_length": 96,
|
| 31 |
"prediction_length": 24
|
| 32 |
},
|
| 33 |
{
|
| 34 |
-
"dataset": "
|
| 35 |
-
"data_fetched_at": "2026-05-
|
| 36 |
"context_length": 96,
|
| 37 |
"prediction_length": 24
|
| 38 |
},
|
| 39 |
{
|
| 40 |
-
"dataset": "
|
| 41 |
-
"data_fetched_at": "2026-05-
|
| 42 |
"context_length": 96,
|
| 43 |
"prediction_length": 24
|
| 44 |
},
|
| 45 |
{
|
| 46 |
-
"dataset": "
|
| 47 |
-
"data_fetched_at": "2026-05-
|
| 48 |
"context_length": 96,
|
| 49 |
"prediction_length": 24
|
| 50 |
},
|
| 51 |
{
|
| 52 |
-
"dataset": "
|
| 53 |
-
"data_fetched_at": "2026-05-
|
| 54 |
"context_length": 96,
|
| 55 |
"prediction_length": 24
|
| 56 |
},
|
| 57 |
{
|
| 58 |
-
"dataset": "
|
| 59 |
-
"data_fetched_at": "2026-05-
|
| 60 |
"context_length": 96,
|
| 61 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 62 |
}
|
| 63 |
]
|
| 64 |
},
|
| 65 |
"Chronos-Bolt-Mini": {
|
| 66 |
"model": "Chronos-Bolt-Mini",
|
| 67 |
"api_model_id": "amazon/chronos-bolt-mini",
|
| 68 |
-
"evaluated_at": "2026-05-
|
| 69 |
"datasets": [
|
| 70 |
{
|
| 71 |
-
"dataset": "
|
| 72 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 73 |
"context_length": 96,
|
| 74 |
"prediction_length": 24
|
| 75 |
},
|
| 76 |
{
|
| 77 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 78 |
-
"data_fetched_at": "2026-05-
|
| 79 |
-
"context_length":
|
| 80 |
"prediction_length": 24
|
| 81 |
},
|
| 82 |
{
|
| 83 |
-
"dataset": "
|
| 84 |
-
"data_fetched_at": "2026-05-
|
| 85 |
"context_length": 96,
|
| 86 |
"prediction_length": 24
|
| 87 |
},
|
| 88 |
{
|
| 89 |
-
"dataset": "
|
| 90 |
-
"data_fetched_at": "2026-05-
|
| 91 |
"context_length": 96,
|
| 92 |
"prediction_length": 24
|
| 93 |
},
|
| 94 |
{
|
| 95 |
-
"dataset": "
|
| 96 |
-
"data_fetched_at": "2026-05-
|
| 97 |
"context_length": 96,
|
| 98 |
"prediction_length": 24
|
| 99 |
},
|
| 100 |
{
|
| 101 |
-
"dataset": "
|
| 102 |
-
"data_fetched_at": "2026-05-
|
| 103 |
"context_length": 96,
|
| 104 |
"prediction_length": 24
|
| 105 |
},
|
| 106 |
{
|
| 107 |
-
"dataset": "
|
| 108 |
-
"data_fetched_at": "2026-05-
|
| 109 |
"context_length": 96,
|
| 110 |
"prediction_length": 24
|
| 111 |
},
|
| 112 |
{
|
| 113 |
-
"dataset": "
|
| 114 |
-
"data_fetched_at": "2026-05-
|
| 115 |
"context_length": 96,
|
| 116 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 117 |
}
|
| 118 |
]
|
| 119 |
},
|
| 120 |
"Chronos-Bolt-Base": {
|
| 121 |
"model": "Chronos-Bolt-Base",
|
| 122 |
"api_model_id": "amazon/chronos-bolt-base",
|
| 123 |
-
"evaluated_at": "2026-05-
|
| 124 |
"datasets": [
|
| 125 |
{
|
| 126 |
-
"dataset": "
|
| 127 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 128 |
"context_length": 96,
|
| 129 |
"prediction_length": 24
|
| 130 |
},
|
| 131 |
{
|
| 132 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 133 |
-
"data_fetched_at": "2026-05-
|
| 134 |
-
"context_length":
|
| 135 |
"prediction_length": 24
|
| 136 |
},
|
| 137 |
{
|
| 138 |
-
"dataset": "
|
| 139 |
-
"data_fetched_at": "2026-05-
|
| 140 |
"context_length": 96,
|
| 141 |
"prediction_length": 24
|
| 142 |
},
|
| 143 |
{
|
| 144 |
-
"dataset": "
|
| 145 |
-
"data_fetched_at": "2026-05-
|
| 146 |
"context_length": 96,
|
| 147 |
"prediction_length": 24
|
| 148 |
},
|
| 149 |
{
|
| 150 |
-
"dataset": "
|
| 151 |
-
"data_fetched_at": "2026-05-
|
| 152 |
"context_length": 96,
|
| 153 |
"prediction_length": 24
|
| 154 |
},
|
| 155 |
{
|
| 156 |
-
"dataset": "
|
| 157 |
-
"data_fetched_at": "2026-05-
|
| 158 |
"context_length": 96,
|
| 159 |
"prediction_length": 24
|
| 160 |
},
|
| 161 |
{
|
| 162 |
-
"dataset": "
|
| 163 |
-
"data_fetched_at": "2026-05-
|
| 164 |
"context_length": 96,
|
| 165 |
"prediction_length": 24
|
| 166 |
},
|
| 167 |
{
|
| 168 |
-
"dataset": "
|
| 169 |
-
"data_fetched_at": "2026-05-
|
| 170 |
"context_length": 96,
|
| 171 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 172 |
}
|
| 173 |
]
|
| 174 |
},
|
| 175 |
"Chronos-2": {
|
| 176 |
"model": "Chronos-2",
|
| 177 |
"api_model_id": "amazon/chronos-2",
|
| 178 |
-
"evaluated_at": "2026-05-
|
| 179 |
"datasets": [
|
| 180 |
{
|
| 181 |
-
"dataset": "
|
| 182 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 183 |
"context_length": 96,
|
| 184 |
"prediction_length": 24
|
| 185 |
},
|
| 186 |
{
|
| 187 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 188 |
-
"data_fetched_at": "2026-05-
|
| 189 |
-
"context_length":
|
| 190 |
"prediction_length": 24
|
| 191 |
},
|
| 192 |
{
|
| 193 |
-
"dataset": "
|
| 194 |
-
"data_fetched_at": "2026-05-
|
| 195 |
"context_length": 96,
|
| 196 |
"prediction_length": 24
|
| 197 |
},
|
| 198 |
{
|
| 199 |
-
"dataset": "
|
| 200 |
-
"data_fetched_at": "2026-05-
|
| 201 |
"context_length": 96,
|
| 202 |
"prediction_length": 24
|
| 203 |
},
|
| 204 |
{
|
| 205 |
-
"dataset": "
|
| 206 |
-
"data_fetched_at": "2026-05-
|
| 207 |
"context_length": 96,
|
| 208 |
"prediction_length": 24
|
| 209 |
},
|
| 210 |
{
|
| 211 |
-
"dataset": "
|
| 212 |
-
"data_fetched_at": "2026-05-
|
| 213 |
"context_length": 96,
|
| 214 |
"prediction_length": 24
|
| 215 |
},
|
| 216 |
{
|
| 217 |
-
"dataset": "
|
| 218 |
-
"data_fetched_at": "2026-05-
|
| 219 |
"context_length": 96,
|
| 220 |
"prediction_length": 24
|
| 221 |
},
|
| 222 |
{
|
| 223 |
-
"dataset": "
|
| 224 |
-
"data_fetched_at": "2026-05-
|
| 225 |
"context_length": 96,
|
| 226 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 227 |
}
|
| 228 |
]
|
| 229 |
},
|
| 230 |
"TimesFM-2.5": {
|
| 231 |
"model": "TimesFM-2.5",
|
| 232 |
"api_model_id": "google/timesfm-2.5-200m-pytorch",
|
| 233 |
-
"evaluated_at": "2026-05-
|
| 234 |
"datasets": [
|
| 235 |
{
|
| 236 |
-
"dataset": "
|
| 237 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 238 |
"context_length": 96,
|
| 239 |
"prediction_length": 24
|
| 240 |
},
|
| 241 |
{
|
| 242 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 243 |
-
"data_fetched_at": "2026-05-
|
| 244 |
-
"context_length":
|
| 245 |
"prediction_length": 24
|
| 246 |
},
|
| 247 |
{
|
| 248 |
-
"dataset": "
|
| 249 |
-
"data_fetched_at": "2026-05-
|
| 250 |
"context_length": 96,
|
| 251 |
"prediction_length": 24
|
| 252 |
},
|
| 253 |
{
|
| 254 |
-
"dataset": "
|
| 255 |
-
"data_fetched_at": "2026-05-
|
| 256 |
"context_length": 96,
|
| 257 |
"prediction_length": 24
|
| 258 |
},
|
| 259 |
{
|
| 260 |
-
"dataset": "
|
| 261 |
-
"data_fetched_at": "2026-05-
|
| 262 |
"context_length": 96,
|
| 263 |
"prediction_length": 24
|
| 264 |
},
|
| 265 |
{
|
| 266 |
-
"dataset": "
|
| 267 |
-
"data_fetched_at": "2026-05-
|
| 268 |
"context_length": 96,
|
| 269 |
"prediction_length": 24
|
| 270 |
},
|
| 271 |
{
|
| 272 |
-
"dataset": "
|
| 273 |
-
"data_fetched_at": "2026-05-
|
| 274 |
"context_length": 96,
|
| 275 |
"prediction_length": 24
|
| 276 |
},
|
| 277 |
{
|
| 278 |
-
"dataset": "
|
| 279 |
-
"data_fetched_at": "2026-05-
|
| 280 |
"context_length": 96,
|
| 281 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 282 |
}
|
| 283 |
]
|
| 284 |
},
|
| 285 |
"Moirai-1.1-Small": {
|
| 286 |
"model": "Moirai-1.1-Small",
|
| 287 |
"api_model_id": "Salesforce/moirai-1.1-R-small",
|
| 288 |
-
"evaluated_at": "2026-05-
|
| 289 |
"datasets": [
|
| 290 |
{
|
| 291 |
-
"dataset": "
|
| 292 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 293 |
"context_length": 96,
|
| 294 |
"prediction_length": 24
|
| 295 |
},
|
| 296 |
{
|
| 297 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 298 |
-
"data_fetched_at": "2026-05-
|
| 299 |
-
"context_length":
|
| 300 |
"prediction_length": 24
|
| 301 |
},
|
| 302 |
{
|
| 303 |
-
"dataset": "
|
| 304 |
-
"data_fetched_at": "2026-05-
|
| 305 |
"context_length": 96,
|
| 306 |
"prediction_length": 24
|
| 307 |
},
|
| 308 |
{
|
| 309 |
-
"dataset": "
|
| 310 |
-
"data_fetched_at": "2026-05-
|
| 311 |
"context_length": 96,
|
| 312 |
"prediction_length": 24
|
| 313 |
},
|
| 314 |
{
|
| 315 |
-
"dataset": "
|
| 316 |
-
"data_fetched_at": "2026-05-
|
| 317 |
"context_length": 96,
|
| 318 |
"prediction_length": 24
|
| 319 |
},
|
| 320 |
{
|
| 321 |
-
"dataset": "
|
| 322 |
-
"data_fetched_at": "2026-05-
|
| 323 |
"context_length": 96,
|
| 324 |
"prediction_length": 24
|
| 325 |
},
|
| 326 |
{
|
| 327 |
-
"dataset": "
|
| 328 |
-
"data_fetched_at": "2026-05-
|
| 329 |
"context_length": 96,
|
| 330 |
"prediction_length": 24
|
| 331 |
},
|
| 332 |
{
|
| 333 |
-
"dataset": "
|
| 334 |
-
"data_fetched_at": "2026-05-
|
| 335 |
"context_length": 96,
|
| 336 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 337 |
}
|
| 338 |
]
|
| 339 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"status": "ok",
|
| 3 |
+
"started_at": "2026-05-28T12:16:39.312039+00:00",
|
| 4 |
+
"finished_at": "2026-05-28T12:21:49.773564+00:00",
|
| 5 |
"data_source": "ts_bench",
|
| 6 |
"data_config": "configs/datasets/ts_bench.yaml",
|
| 7 |
"ts_bench_root": "/Users/zhouziyu/Desktop/TS-Bench",
|
|
|
|
| 10 |
"Chronos-Bolt-Tiny": {
|
| 11 |
"model": "Chronos-Bolt-Tiny",
|
| 12 |
"api_model_id": "amazon/chronos-bolt-tiny",
|
| 13 |
+
"evaluated_at": "2026-05-28T12:18:05.053451+00:00",
|
| 14 |
"datasets": [
|
| 15 |
{
|
| 16 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 17 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 18 |
+
"context_length": 96,
|
| 19 |
+
"prediction_length": 24
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 23 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 24 |
"context_length": 96,
|
| 25 |
"prediction_length": 24
|
| 26 |
},
|
| 27 |
{
|
| 28 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 29 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 30 |
+
"context_length": 83,
|
| 31 |
"prediction_length": 24
|
| 32 |
},
|
| 33 |
{
|
| 34 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 35 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 36 |
"context_length": 96,
|
| 37 |
"prediction_length": 24
|
| 38 |
},
|
| 39 |
{
|
| 40 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 41 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 42 |
"context_length": 96,
|
| 43 |
"prediction_length": 24
|
| 44 |
},
|
| 45 |
{
|
| 46 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 47 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 48 |
"context_length": 96,
|
| 49 |
"prediction_length": 24
|
| 50 |
},
|
| 51 |
{
|
| 52 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 53 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 54 |
"context_length": 96,
|
| 55 |
"prediction_length": 24
|
| 56 |
},
|
| 57 |
{
|
| 58 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 59 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 60 |
"context_length": 96,
|
| 61 |
"prediction_length": 24
|
| 62 |
},
|
| 63 |
{
|
| 64 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 65 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 66 |
"context_length": 96,
|
| 67 |
"prediction_length": 24
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 71 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 72 |
+
"context_length": 4,
|
| 73 |
+
"prediction_length": 2
|
| 74 |
}
|
| 75 |
]
|
| 76 |
},
|
| 77 |
"Chronos-Bolt-Mini": {
|
| 78 |
"model": "Chronos-Bolt-Mini",
|
| 79 |
"api_model_id": "amazon/chronos-bolt-mini",
|
| 80 |
+
"evaluated_at": "2026-05-28T12:18:55.358936+00:00",
|
| 81 |
"datasets": [
|
| 82 |
{
|
| 83 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 84 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 85 |
+
"context_length": 96,
|
| 86 |
+
"prediction_length": 24
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 90 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 91 |
"context_length": 96,
|
| 92 |
"prediction_length": 24
|
| 93 |
},
|
| 94 |
{
|
| 95 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 96 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 97 |
+
"context_length": 83,
|
| 98 |
"prediction_length": 24
|
| 99 |
},
|
| 100 |
{
|
| 101 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 102 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 103 |
"context_length": 96,
|
| 104 |
"prediction_length": 24
|
| 105 |
},
|
| 106 |
{
|
| 107 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 108 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 109 |
"context_length": 96,
|
| 110 |
"prediction_length": 24
|
| 111 |
},
|
| 112 |
{
|
| 113 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 114 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 115 |
"context_length": 96,
|
| 116 |
"prediction_length": 24
|
| 117 |
},
|
| 118 |
{
|
| 119 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 120 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 121 |
"context_length": 96,
|
| 122 |
"prediction_length": 24
|
| 123 |
},
|
| 124 |
{
|
| 125 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 126 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 127 |
"context_length": 96,
|
| 128 |
"prediction_length": 24
|
| 129 |
},
|
| 130 |
{
|
| 131 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 132 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 133 |
"context_length": 96,
|
| 134 |
"prediction_length": 24
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 138 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 139 |
+
"context_length": 4,
|
| 140 |
+
"prediction_length": 2
|
| 141 |
}
|
| 142 |
]
|
| 143 |
},
|
| 144 |
"Chronos-Bolt-Base": {
|
| 145 |
"model": "Chronos-Bolt-Base",
|
| 146 |
"api_model_id": "amazon/chronos-bolt-base",
|
| 147 |
+
"evaluated_at": "2026-05-28T12:19:28.056546+00:00",
|
| 148 |
"datasets": [
|
| 149 |
{
|
| 150 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 151 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 152 |
+
"context_length": 96,
|
| 153 |
+
"prediction_length": 24
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 157 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 158 |
"context_length": 96,
|
| 159 |
"prediction_length": 24
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 163 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 164 |
+
"context_length": 83,
|
| 165 |
"prediction_length": 24
|
| 166 |
},
|
| 167 |
{
|
| 168 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 169 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 170 |
"context_length": 96,
|
| 171 |
"prediction_length": 24
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 175 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 176 |
"context_length": 96,
|
| 177 |
"prediction_length": 24
|
| 178 |
},
|
| 179 |
{
|
| 180 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 181 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 182 |
"context_length": 96,
|
| 183 |
"prediction_length": 24
|
| 184 |
},
|
| 185 |
{
|
| 186 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 187 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 188 |
"context_length": 96,
|
| 189 |
"prediction_length": 24
|
| 190 |
},
|
| 191 |
{
|
| 192 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 193 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 194 |
"context_length": 96,
|
| 195 |
"prediction_length": 24
|
| 196 |
},
|
| 197 |
{
|
| 198 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 199 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 200 |
"context_length": 96,
|
| 201 |
"prediction_length": 24
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 205 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 206 |
+
"context_length": 4,
|
| 207 |
+
"prediction_length": 2
|
| 208 |
}
|
| 209 |
]
|
| 210 |
},
|
| 211 |
"Chronos-2": {
|
| 212 |
"model": "Chronos-2",
|
| 213 |
"api_model_id": "amazon/chronos-2",
|
| 214 |
+
"evaluated_at": "2026-05-28T12:20:18.799459+00:00",
|
| 215 |
"datasets": [
|
| 216 |
{
|
| 217 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 218 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 219 |
+
"context_length": 96,
|
| 220 |
+
"prediction_length": 24
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 224 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 225 |
"context_length": 96,
|
| 226 |
"prediction_length": 24
|
| 227 |
},
|
| 228 |
{
|
| 229 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 230 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 231 |
+
"context_length": 83,
|
| 232 |
"prediction_length": 24
|
| 233 |
},
|
| 234 |
{
|
| 235 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 236 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 237 |
"context_length": 96,
|
| 238 |
"prediction_length": 24
|
| 239 |
},
|
| 240 |
{
|
| 241 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 242 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 243 |
"context_length": 96,
|
| 244 |
"prediction_length": 24
|
| 245 |
},
|
| 246 |
{
|
| 247 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 248 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 249 |
"context_length": 96,
|
| 250 |
"prediction_length": 24
|
| 251 |
},
|
| 252 |
{
|
| 253 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 254 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 255 |
"context_length": 96,
|
| 256 |
"prediction_length": 24
|
| 257 |
},
|
| 258 |
{
|
| 259 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 260 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 261 |
"context_length": 96,
|
| 262 |
"prediction_length": 24
|
| 263 |
},
|
| 264 |
{
|
| 265 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 266 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 267 |
"context_length": 96,
|
| 268 |
"prediction_length": 24
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 272 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 273 |
+
"context_length": 4,
|
| 274 |
+
"prediction_length": 2
|
| 275 |
}
|
| 276 |
]
|
| 277 |
},
|
| 278 |
"TimesFM-2.5": {
|
| 279 |
"model": "TimesFM-2.5",
|
| 280 |
"api_model_id": "google/timesfm-2.5-200m-pytorch",
|
| 281 |
+
"evaluated_at": "2026-05-28T12:21:19.114949+00:00",
|
| 282 |
"datasets": [
|
| 283 |
{
|
| 284 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 285 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 286 |
+
"context_length": 96,
|
| 287 |
+
"prediction_length": 24
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 291 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 292 |
"context_length": 96,
|
| 293 |
"prediction_length": 24
|
| 294 |
},
|
| 295 |
{
|
| 296 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 297 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 298 |
+
"context_length": 83,
|
| 299 |
"prediction_length": 24
|
| 300 |
},
|
| 301 |
{
|
| 302 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 303 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 304 |
"context_length": 96,
|
| 305 |
"prediction_length": 24
|
| 306 |
},
|
| 307 |
{
|
| 308 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 309 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 310 |
"context_length": 96,
|
| 311 |
"prediction_length": 24
|
| 312 |
},
|
| 313 |
{
|
| 314 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 315 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 316 |
"context_length": 96,
|
| 317 |
"prediction_length": 24
|
| 318 |
},
|
| 319 |
{
|
| 320 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 321 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 322 |
"context_length": 96,
|
| 323 |
"prediction_length": 24
|
| 324 |
},
|
| 325 |
{
|
| 326 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 327 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 328 |
"context_length": 96,
|
| 329 |
"prediction_length": 24
|
| 330 |
},
|
| 331 |
{
|
| 332 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 333 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 334 |
"context_length": 96,
|
| 335 |
"prediction_length": 24
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 339 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 340 |
+
"context_length": 4,
|
| 341 |
+
"prediction_length": 2
|
| 342 |
}
|
| 343 |
]
|
| 344 |
},
|
| 345 |
"Moirai-1.1-Small": {
|
| 346 |
"model": "Moirai-1.1-Small",
|
| 347 |
"api_model_id": "Salesforce/moirai-1.1-R-small",
|
| 348 |
+
"evaluated_at": "2026-05-28T12:21:49.771336+00:00",
|
| 349 |
"datasets": [
|
| 350 |
{
|
| 351 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 352 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 353 |
+
"context_length": 96,
|
| 354 |
+
"prediction_length": 24
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 358 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 359 |
"context_length": 96,
|
| 360 |
"prediction_length": 24
|
| 361 |
},
|
| 362 |
{
|
| 363 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 364 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 365 |
+
"context_length": 83,
|
| 366 |
"prediction_length": 24
|
| 367 |
},
|
| 368 |
{
|
| 369 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 370 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 371 |
"context_length": 96,
|
| 372 |
"prediction_length": 24
|
| 373 |
},
|
| 374 |
{
|
| 375 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 376 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 377 |
"context_length": 96,
|
| 378 |
"prediction_length": 24
|
| 379 |
},
|
| 380 |
{
|
| 381 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 382 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 383 |
"context_length": 96,
|
| 384 |
"prediction_length": 24
|
| 385 |
},
|
| 386 |
{
|
| 387 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 388 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 389 |
"context_length": 96,
|
| 390 |
"prediction_length": 24
|
| 391 |
},
|
| 392 |
{
|
| 393 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 394 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 395 |
"context_length": 96,
|
| 396 |
"prediction_length": 24
|
| 397 |
},
|
| 398 |
{
|
| 399 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 400 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 401 |
"context_length": 96,
|
| 402 |
"prediction_length": 24
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 406 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 407 |
+
"context_length": 4,
|
| 408 |
+
"prediction_length": 2
|
| 409 |
}
|
| 410 |
]
|
| 411 |
}
|
results/timesfm_2_5/all_results.csv
CHANGED
|
@@ -1,9 +1,11 @@
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
-
binance_btcusdt_1h_close/1h/short,TimesFM-2.5,
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
|
| 8 |
-
|
| 9 |
-
|
|
|
|
|
|
|
|
|
| 1 |
dataset,model,eval_metrics/MSE[mean],eval_metrics/MSE[0.5],eval_metrics/MAE[0.5],eval_metrics/MASE[0.5],eval_metrics/MAPE[0.5],eval_metrics/sMAPE[0.5],eval_metrics/MSIS,eval_metrics/RMSE[mean],eval_metrics/NRMSE[mean],eval_metrics/ND[0.5],eval_metrics/mean_weighted_sum_quantile_loss,domain,num_variates
|
| 2 |
+
binance_btcusdt_1h_close/1h/short,TimesFM-2.5,3184300.868250529,2877341.8473765054,1502.9791666666667,1.9366630929550903,0.020371679342378764,0.0201094796089761,86.07571948209704,1784.4609461264567,0.024049315263214986,0.02025576401191205,0.021796525719852494,Finance,1
|
| 3 |
+
coingecko_bitcoin_7d_price_usd/1h/short,TimesFM-2.5,2978472.513104757,2691326.2771759033,1431.572265625,1.7855091583102543,0.01942591517249738,0.019179939216192834,78.84789791391036,1725.8251687539955,0.023269242362690887,0.01930184042488197,0.020704134741179717,Finance,1
|
| 4 |
+
nasa_power_shanghai_hourly_T2M/1h/short,TimesFM-2.5,7.424815583824056,11.590662498969019,2.684919834136963,0.8565959528394198,0.09649247329369882,0.10426336051909445,5.473512568019115,2.7248514792230525,0.1061647697821695,0.10460896612097444,0.06960420118115515,Weather,1
|
| 5 |
+
noaa_coops_sf_water_level_water_level/6min/short,TimesFM-2.5,0.014191201840556963,0.008006066655363927,0.08215828488270442,3.0984664146014134,0.37820721429781673,0.3032819543290635,207.6676558778752,0.11912683090117425,0.4636788830559884,0.31978573995484816,0.46509102471709085,Ocean,1
|
| 6 |
+
noaa_ndbc_46013_realtime_wave_height/10min/short,TimesFM-2.5,0.05512604274710261,0.031698439646326904,0.15231487154960632,1.09620529318399,0.06015022141194603,0.05812306444413484,72.78888463699143,0.23478935824926694,0.09001509161876256,0.05839547933378565,0.08648648191844961,Ocean,1
|
| 7 |
+
nws_ksfo_observations_temperature/1h/short,TimesFM-2.5,0.0034255919382151965,0.0033221443641195947,0.0194094975789388,0.36775869844608894,0.0015129974061875602,0.0015030149122856155,15.604036901152464,0.05852855660457719,0.0045079760301519585,0.0014949548548461198,0.0016153432104989958,Weather,1
|
| 8 |
+
open_meteo_shanghai_air_quality_hourly_pm2_5/1h/short,TimesFM-2.5,128.3803764339212,199.8811043637435,9.946486552556356,0.5349970940929124,0.19267528355159025,0.23408765568966516,7.410677562806579,11.330506450901531,0.2308618296844907,0.20266208703077312,0.166486311960706,Air Quality,1
|
| 9 |
+
open_meteo_shanghai_hourly_temperature_2m/1h/short,TimesFM-2.5,0.4488093384068937,0.42550756879063556,0.5227727890014648,0.28471744342993616,0.021116368406687125,0.020967622620989168,11.867516613268592,0.6699323386782382,0.026734911000282555,0.020862232169443492,0.02480197235359902,Weather,1
|
| 10 |
+
usgs_potomac_iv_usgs_00060/15min/short,TimesFM-2.5,3964.9586008389792,2807.6309466362,52.009847005208336,nan,0.002045911466877187,0.0020444056954015345,nan,62.9679172344058,0.0024762084444137954,0.002045283185523513,0.002015193888024105,Hydrology,1
|
| 11 |
+
wikimedia_time_series_hourly_pageviews/1D/short,TimesFM-2.5,4207.680999754579,5861.192346271244,75.43997955322266,1.6165709904262,0.24218997554725216,0.2765631520841845,22.796233912876673,64.86664011458107,0.20891027412103405,0.2429628971118282,0.177518524153541,Web,1
|
results/timesfm_2_5/online_meta.json
CHANGED
|
@@ -1,55 +1,67 @@
|
|
| 1 |
{
|
| 2 |
"model": "TimesFM-2.5",
|
| 3 |
"api_model_id": "google/timesfm-2.5-200m-pytorch",
|
| 4 |
-
"evaluated_at": "2026-05-
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
-
"dataset": "
|
| 8 |
-
"data_fetched_at": "2026-05-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 9 |
"context_length": 96,
|
| 10 |
"prediction_length": 24
|
| 11 |
},
|
| 12 |
{
|
| 13 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 14 |
-
"data_fetched_at": "2026-05-
|
| 15 |
-
"context_length":
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
-
"dataset": "
|
| 20 |
-
"data_fetched_at": "2026-05-
|
| 21 |
"context_length": 96,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
-
"dataset": "
|
| 26 |
-
"data_fetched_at": "2026-05-
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
-
"dataset": "
|
| 32 |
-
"data_fetched_at": "2026-05-
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
-
"dataset": "
|
| 38 |
-
"data_fetched_at": "2026-05-
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
-
"dataset": "
|
| 44 |
-
"data_fetched_at": "2026-05-
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
-
"dataset": "
|
| 50 |
-
"data_fetched_at": "2026-05-
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 53 |
}
|
| 54 |
]
|
| 55 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"model": "TimesFM-2.5",
|
| 3 |
"api_model_id": "google/timesfm-2.5-200m-pytorch",
|
| 4 |
+
"evaluated_at": "2026-05-28T12:21:19.114949+00:00",
|
| 5 |
"datasets": [
|
| 6 |
{
|
| 7 |
+
"dataset": "task_binance_btcusdt_1h_crypto_btcusdt_close_20260527t120000z",
|
| 8 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 9 |
+
"context_length": 96,
|
| 10 |
+
"prediction_length": 24
|
| 11 |
+
},
|
| 12 |
+
{
|
| 13 |
+
"dataset": "task_coingecko_bitcoin_7d_coin_bitcoin_price_usd_20260527t120016z",
|
| 14 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 15 |
"context_length": 96,
|
| 16 |
"prediction_length": 24
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"dataset": "task_nasa_power_shanghai_hourly_point_shanghai_31.2304_121.4737_t2m_20260524t230000z",
|
| 20 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 21 |
+
"context_length": 83,
|
| 22 |
"prediction_length": 24
|
| 23 |
},
|
| 24 |
{
|
| 25 |
+
"dataset": "task_noaa_coops_sf_water_level_coops_station_9414290_water_level_20260528t094200z",
|
| 26 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 27 |
"context_length": 96,
|
| 28 |
"prediction_length": 24
|
| 29 |
},
|
| 30 |
{
|
| 31 |
+
"dataset": "task_noaa_ndbc_46013_realtime_ndbc_station_46013_wave_height_20260528t045000z",
|
| 32 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 33 |
"context_length": 96,
|
| 34 |
"prediction_length": 24
|
| 35 |
},
|
| 36 |
{
|
| 37 |
+
"dataset": "task_nws_ksfo_observations_nws_station_ksfo_temperature_20260528t101000z",
|
| 38 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 39 |
"context_length": 96,
|
| 40 |
"prediction_length": 24
|
| 41 |
},
|
| 42 |
{
|
| 43 |
+
"dataset": "task_open_meteo_shanghai_air_quality_hourly_grid_shanghai_31.2304_121.4737_pm2_5_20260527t120000z",
|
| 44 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 45 |
"context_length": 96,
|
| 46 |
"prediction_length": 24
|
| 47 |
},
|
| 48 |
{
|
| 49 |
+
"dataset": "task_open_meteo_shanghai_hourly_grid_shanghai_31.2304_121.4737_temperature_2m_20260527t120000z",
|
| 50 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 51 |
"context_length": 96,
|
| 52 |
"prediction_length": 24
|
| 53 |
},
|
| 54 |
{
|
| 55 |
+
"dataset": "task_usgs_potomac_iv_usgs_site_01646500_usgs_00060_20260528t095000z",
|
| 56 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 57 |
"context_length": 96,
|
| 58 |
"prediction_length": 24
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"dataset": "task_wikimedia_time_series_hourly_enwiki_article_time_series_pageviews_20260525t000000z",
|
| 62 |
+
"data_fetched_at": "2026-05-28T12:16:39.187896Z",
|
| 63 |
+
"context_length": 4,
|
| 64 |
+
"prediction_length": 2
|
| 65 |
}
|
| 66 |
]
|
| 67 |
}
|
src/about.py
CHANGED
|
@@ -1,11 +1,11 @@
|
|
| 1 |
TITLE = """ TSFM Realworld Bench Leaderboard """
|
| 2 |
|
| 3 |
INTRODUCTION_TEXT = """
|
| 4 |
-
**TSFM Realworld Bench** evaluates time series foundation models on **live
|
| 5 |
-
|
| 6 |
-
|
| 7 |
-
**metric values**
|
| 8 |
-
|
| 9 |
"""
|
| 10 |
|
| 11 |
LLM_BENCHMARKS_TEXT = """
|
|
@@ -18,13 +18,20 @@ LLM_BENCHMARKS_TEXT = """
|
|
| 18 |
|
| 19 |
## Metrics
|
| 20 |
|
| 21 |
-
- **MASE** — Mean Absolute Scaled Error
|
| 22 |
- **CRPS** — Mean Weighted Sum Quantile Loss (probabilistic forecast quality)
|
| 23 |
-
- **CRPS_Rank** / **MASE_Rank** — average per-dataset rank (lower is better)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 24 |
|
| 25 |
## Data interface
|
| 26 |
|
| 27 |
-
Live
|
| 28 |
holds out the most recent forecast horizon as ground truth and calls TSFM models through
|
| 29 |
the **TSFM.ai** API for zero-shot forecasts.
|
| 30 |
|
|
|
|
| 1 |
TITLE = """ TSFM Realworld Bench Leaderboard """
|
| 2 |
|
| 3 |
INTRODUCTION_TEXT = """
|
| 4 |
+
**TSFM Realworld Bench** evaluates time series foundation models on **live TS-Bench
|
| 5 |
+
data** with **zero-shot API inference** via [TSFM.ai](https://tsfm.ai/).
|
| 6 |
+
Following [GIFT-Eval](https://github.com/SalesforceAIResearch/gift-eval), each model's
|
| 7 |
+
**raw metric values** are aggregated with a **geometric mean** across datasets.
|
| 8 |
+
Per-dataset **ranks** are averaged for ranking (lower is better).
|
| 9 |
"""
|
| 10 |
|
| 11 |
LLM_BENCHMARKS_TEXT = """
|
|
|
|
| 18 |
|
| 19 |
## Metrics
|
| 20 |
|
| 21 |
+
- **MASE** — Mean Absolute Scaled Error
|
| 22 |
- **CRPS** — Mean Weighted Sum Quantile Loss (probabilistic forecast quality)
|
| 23 |
+
- **CRPS_Rank** / **MASE_Rank** — average per-dataset rank on raw metrics (lower is better)
|
| 24 |
+
|
| 25 |
+
## Aggregation (GIFT-Eval style)
|
| 26 |
+
|
| 27 |
+
1. Compute evaluation metrics on each dataset configuration (raw values).
|
| 28 |
+
2. Rank models per dataset on each metric.
|
| 29 |
+
3. Aggregate metric values with a **geometric mean** across datasets.
|
| 30 |
+
4. Aggregate ranks with an **arithmetic mean** across datasets.
|
| 31 |
|
| 32 |
## Data interface
|
| 33 |
|
| 34 |
+
Live series are collected from **TS-Bench** (`TsBenchDataSource`). Each evaluation
|
| 35 |
holds out the most recent forecast horizon as ground truth and calls TSFM models through
|
| 36 |
the **TSFM.ai** API for zero-shot forecasts.
|
| 37 |
|
src/utils.py
CHANGED
|
@@ -1,4 +1,3 @@
|
|
| 1 |
-
import json
|
| 2 |
import os
|
| 3 |
import re
|
| 4 |
from pathlib import Path
|
|
@@ -78,37 +77,14 @@ def unify_freq(df: pd.DataFrame) -> pd.DataFrame:
|
|
| 78 |
return df
|
| 79 |
|
| 80 |
|
| 81 |
-
def
|
| 82 |
-
|
| 83 |
-
if status_path.exists():
|
| 84 |
-
baseline = json.loads(status_path.read_text()).get("baseline_model")
|
| 85 |
-
if baseline:
|
| 86 |
-
return baseline
|
| 87 |
-
return "Chronos-Bolt-Base"
|
| 88 |
-
|
| 89 |
-
|
| 90 |
-
def standardize_df(df: pd.DataFrame, root_dir: str = "results") -> pd.DataFrame:
|
| 91 |
df = df.copy()
|
| 92 |
df["model"] = df["model"].replace(LEGACY_MODEL_NAMES)
|
| 93 |
|
| 94 |
for metric in METRIC_COLUMNS:
|
| 95 |
df[metric] = pd.to_numeric(df[metric], errors="coerce")
|
| 96 |
df[metric] = df[metric].fillna(df[metric].mean())
|
| 97 |
-
|
| 98 |
-
baseline_name = load_baseline_name(root_dir)
|
| 99 |
-
if baseline_name not in set(df["model"]):
|
| 100 |
-
return df
|
| 101 |
-
|
| 102 |
-
group_keys = ["dataset", "term_length", "frequency"]
|
| 103 |
-
for _, group in df.groupby(group_keys):
|
| 104 |
-
baseline = group[group["model"] == baseline_name]
|
| 105 |
-
if baseline.empty:
|
| 106 |
-
continue
|
| 107 |
-
idx = baseline.index[0]
|
| 108 |
-
for metric in METRIC_COLUMNS:
|
| 109 |
-
denom = baseline.at[idx, metric]
|
| 110 |
-
if denom and denom != 0:
|
| 111 |
-
df.loc[group.index, metric] = df.loc[group.index, metric] / denom
|
| 112 |
return df
|
| 113 |
|
| 114 |
|
|
@@ -161,7 +137,7 @@ def get_grouped_dfs(
|
|
| 161 |
|
| 162 |
df = load_all_results(root_dir)
|
| 163 |
df = enrich_with_properties(df, ds_properties)
|
| 164 |
-
df =
|
| 165 |
df = _add_per_dataset_ranks(df)
|
| 166 |
|
| 167 |
rank_cols = [f"Rank_{metric}" for metric in DISPLAY_METRICS]
|
|
|
|
|
|
|
| 1 |
import os
|
| 2 |
import re
|
| 3 |
from pathlib import Path
|
|
|
|
| 77 |
return df
|
| 78 |
|
| 79 |
|
| 80 |
+
def prepare_metrics_df(df: pd.DataFrame) -> pd.DataFrame:
|
| 81 |
+
"""Clean metric columns for aggregation (GIFT-Eval style, no model normalization)."""
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 82 |
df = df.copy()
|
| 83 |
df["model"] = df["model"].replace(LEGACY_MODEL_NAMES)
|
| 84 |
|
| 85 |
for metric in METRIC_COLUMNS:
|
| 86 |
df[metric] = pd.to_numeric(df[metric], errors="coerce")
|
| 87 |
df[metric] = df[metric].fillna(df[metric].mean())
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 88 |
return df
|
| 89 |
|
| 90 |
|
|
|
|
| 137 |
|
| 138 |
df = load_all_results(root_dir)
|
| 139 |
df = enrich_with_properties(df, ds_properties)
|
| 140 |
+
df = prepare_metrics_df(df)
|
| 141 |
df = _add_per_dataset_ranks(df)
|
| 142 |
|
| 143 |
rank_cols = [f"Rank_{metric}" for metric in DISPLAY_METRICS]
|