Melmaphother commited on
Commit
9ecdf07
·
verified ·
1 Parent(s): ffb54d0

Add files using upload-large-folder tool

Browse files
Files changed (50) hide show
  1. README.md +48 -47
  2. calibration/bucket_class_biases.csv +121 -0
  3. manifest.json +1652 -0
  4. models/pre/models.json +8 -0
  5. test/predicted_features/flowstate/bitbrains_fast_storage_5T_long/predicted_window_features.npz +3 -0
  6. test/predicted_features/flowstate/bitbrains_fast_storage_5T_medium/predicted_window_features.npz +3 -0
  7. test/predicted_features/flowstate/bitbrains_fast_storage_5T_short/predicted_window_features.npz +3 -0
  8. test/predicted_features/flowstate/bitbrains_fast_storage_H_short/predicted_window_features.npz +3 -0
  9. test/predicted_features/flowstate/bitbrains_rnd_5T_long/predicted_window_features.npz +3 -0
  10. test/predicted_features/flowstate/bitbrains_rnd_5T_medium/predicted_window_features.npz +3 -0
  11. test/predicted_features/flowstate/bitbrains_rnd_5T_short/predicted_window_features.npz +3 -0
  12. test/predicted_features/flowstate/bitbrains_rnd_H_short/predicted_window_features.npz +3 -0
  13. test/predicted_features/flowstate/bizitobs_application_10S_long/predicted_window_features.npz +3 -0
  14. test/predicted_features/flowstate/bizitobs_application_10S_medium/predicted_window_features.npz +3 -0
  15. test/predicted_features/flowstate/bizitobs_application_10S_short/predicted_window_features.npz +3 -0
  16. test/predicted_features/flowstate/bizitobs_l2c_5T_long/predicted_window_features.npz +3 -0
  17. test/predicted_features/flowstate/bizitobs_l2c_5T_medium/predicted_window_features.npz +3 -0
  18. test/predicted_features/flowstate/bizitobs_l2c_5T_short/predicted_window_features.npz +3 -0
  19. test/predicted_features/flowstate/bizitobs_l2c_H_long/predicted_window_features.npz +3 -0
  20. test/predicted_features/flowstate/bizitobs_l2c_H_medium/predicted_window_features.npz +3 -0
  21. test/predicted_features/flowstate/bizitobs_l2c_H_short/predicted_window_features.npz +3 -0
  22. test/predicted_features/flowstate/bizitobs_service_10S_long/predicted_window_features.npz +3 -0
  23. test/predicted_features/flowstate/bizitobs_service_10S_medium/predicted_window_features.npz +3 -0
  24. test/predicted_features/flowstate/bizitobs_service_10S_short/predicted_window_features.npz +3 -0
  25. test/predicted_features/flowstate/car_parts_M_short/predicted_window_features.npz +3 -0
  26. test/predicted_features/flowstate/covid_deaths_D_short/predicted_window_features.npz +3 -0
  27. test/predicted_features/flowstate/electricity_15T_long/predicted_window_features.npz +3 -0
  28. test/predicted_features/flowstate/electricity_15T_medium/predicted_window_features.npz +3 -0
  29. test/predicted_features/flowstate/electricity_15T_short/predicted_window_features.npz +3 -0
  30. test/predicted_features/flowstate/electricity_D_short/predicted_window_features.npz +3 -0
  31. test/predicted_features/flowstate/electricity_H_long/predicted_window_features.npz +3 -0
  32. test/predicted_features/timesfm-2.5/loop_seattle_H_short/predicted_window_features.npz +3 -0
  33. test/predicted_features/timesfm-2.5/m4_quarterly_Q_short/predicted_window_features.npz +3 -0
  34. test/predicted_features/timesfm-2.5/saugeen_D_short/predicted_window_features.npz +3 -0
  35. test/predicted_features/timesfm-2.5/saugeen_W_short/predicted_window_features.npz +3 -0
  36. test/predicted_features/timesfm-2.5/solar_10T_long/predicted_window_features.npz +3 -0
  37. test/predicted_features/timesfm-2.5/solar_10T_medium/predicted_window_features.npz +3 -0
  38. test/predicted_features/timesfm-2.5/solar_10T_short/predicted_window_features.npz +3 -0
  39. test/predicted_features/timesfm-2.5/solar_D_short/predicted_window_features.npz +3 -0
  40. test/predicted_features/timesfm-2.5/solar_H_long/predicted_window_features.npz +3 -0
  41. test/predicted_features/timesfm-2.5/solar_H_medium/predicted_window_features.npz +3 -0
  42. test/predicted_features/timesfm-2.5/solar_H_short/predicted_window_features.npz +3 -0
  43. test/predicted_features/timesfm-2.5/solar_W_short/predicted_window_features.npz +3 -0
  44. test/predicted_features/timesfm-2.5/sz_taxi_15T_long/predicted_window_features.npz +3 -0
  45. test/predicted_features/timesfm-2.5/sz_taxi_15T_medium/predicted_window_features.npz +3 -0
  46. test/predicted_features/timesfm-2.5/sz_taxi_15T_short/predicted_window_features.npz +3 -0
  47. test/predicted_features/timesfm-2.5/sz_taxi_H_short/predicted_window_features.npz +3 -0
  48. test/predicted_features/timesfm-2.5/us_births_D_short/predicted_window_features.npz +3 -0
  49. test/predicted_features/timesfm-2.5/us_births_M_short/predicted_window_features.npz +3 -0
  50. test/predicted_features/timesfm-2.5/us_births_W_short/predicted_window_features.npz +3 -0
README.md CHANGED
@@ -1,7 +1,7 @@
1
  ---
2
  license: apache-2.0
3
  pipeline_tag: time-series-forecasting
4
- library_name: pytorch
5
  base_model:
6
  - amazon/chronos-2
7
  - google/timesfm-2.5-200m-pytorch
@@ -12,59 +12,66 @@ base_model:
12
  tags:
13
  - time-series-forecasting
14
  - foundation-models
15
- - pretrained-models
16
  - time-series
17
- - timeseries
18
  - forecasting
19
  - ensemble
20
- - meta-learning
21
  - agentic
22
  - gift-eval
23
  - xgboost
24
-
25
  ---
26
 
27
- > [!WARNING]
28
- > **This is a benchmarking artifact for the GIFT-Eval submission.**
29
- > CastStar is an agentic time-series forecasting system built on top of multiple foundation forecasting models. The submitted GIFT-Eval result uses pre-computed forecasts and a meta-selection/ensemble layer over the model pool. It is intended to make the leaderboard submission reproducible and comparable under the GIFT-Eval protocol.
30
-
31
- ## What This Is
32
-
33
- CastStar combines forecasts from a pool of time-series foundation models and uses a learned gating strategy to select or weight base-model predictions for each forecasting setting. The system follows the same general meta-forecasting idea as FFORMA-style model selection, where lightweight time-series features and dataset metadata guide the choice of forecasting experts.
34
 
35
- The GIFT-Eval submission evaluates CastStar on all 97 dataset configurations using the official CSV format.
 
 
 
36
 
37
  ## Model Pool
38
 
39
- CastStar uses the following base forecasting models. Compared with the Toto-2.0-FnF setup, CastStar only uses the **Toto-2.0-2.5B** checkpoint from the Toto family.
 
 
 
 
 
 
 
40
 
41
- | # | Model | Family |
42
- | :--: | ------------------------------------------------------------ | :-------: |
43
- | 0 | [chronos-2](https://huggingface.co/amazon/chronos-2) | Chronos |
44
- | 1 | [timesfm-2.5](https://huggingface.co/google/timesfm-2.5-200m-pytorch) | TimesFM |
45
- | 2 | [flowstate](https://huggingface.co/ibm-research/flowstate) | FlowState |
46
- | 3 | [tirex](https://huggingface.co/NX-AI/TiRex-1.1-gifteval) | TiRex |
47
- | 4 | [patchtst-fm](https://huggingface.co/ibm-research/patchtst-fm-r1) | PatchTST |
48
- | 5 | [toto-2.0-2.5b](https://huggingface.co/Datadog/Toto-2.0-2.5B) | Toto 2.0 |
49
 
50
- ## Key Features
51
 
52
- - **Agentic forecasting system:** CastStar uses multiple forecasting experts and a decision layer to produce final probabilistic forecasts.
53
- - **Foundation-model pool:** The model pool includes Chronos-2, TimesFM-2.5, FlowState, TiRex, PatchTST-FM, and Toto-2.0-2.5B.
54
- - **GIFT-Eval compatible:** Results are submitted in the official GIFT-Eval format with 97/97 dataset configurations and 11 evaluation metrics.
55
- - **No test-data leakage:** The submitted configuration reports no GIFT-Eval test-data leakage.
56
- - **Benchmark-focused artifact:** The submitted result is designed for leaderboard evaluation and reproducibility under the GIFT-Eval protocol.
 
 
 
 
 
 
 
 
 
57
 
58
- ## GIFT-Eval Submission
 
59
 
60
- The submitted files are:
61
 
62
- ```text
63
- results/CastStar/all_results.csv
64
- results/CastStar/config.json
65
- ```
 
 
 
 
66
 
67
- Submission metadata:
68
 
69
  ```json
70
  {
@@ -72,20 +79,14 @@ Submission metadata:
72
  "model_type": "agentic",
73
  "model_dtype": "float32",
74
  "model_link": "https://huggingface.co/USTC-AGI/CastStar",
75
- "code_link": "https://github.com/ustc-time-series/CastStar",
76
- "org": "CastStar",
77
  "testdata_leakage": "No",
78
- "replication_code_available": "No"
79
  }
80
  ```
81
 
82
- ## Additional Resources
83
-
84
- - [GIFT-Eval benchmark](https://huggingface.co/spaces/Salesforce/GIFT-Eval)
85
- - [GIFT-Eval GitHub repository](https://github.com/SalesforceAIResearch/gift-eval)
86
- - [CastStar GitHub repository](https://github.com/ustc-time-series/CastStar)
87
- - [CastStar model page](https://huggingface.co/USTC-AGI/CastStar)
88
-
89
- ## Citation
90
 
91
- If you use CastStar, please cite the corresponding CastStar paper or repository when available.
 
 
1
  ---
2
  license: apache-2.0
3
  pipeline_tag: time-series-forecasting
4
+ library_name: xgboost
5
  base_model:
6
  - amazon/chronos-2
7
  - google/timesfm-2.5-200m-pytorch
 
12
  tags:
13
  - time-series-forecasting
14
  - foundation-models
 
15
  - time-series
 
16
  - forecasting
17
  - ensemble
 
18
  - agentic
19
  - gift-eval
20
  - xgboost
 
21
  ---
22
 
23
+ # CastStar
 
 
 
 
 
 
24
 
25
+ CastStar is an agentic time-series forecasting system built on multiple
26
+ forecasting foundation models. This repository contains the frozen inference
27
+ assets for reproducing the CastStar submission on all 97 GIFT-Eval dataset
28
+ configurations.
29
 
30
  ## Model Pool
31
 
32
+ | Model | Source |
33
+ |---|---|
34
+ | Chronos-2 | [amazon/chronos-2](https://huggingface.co/amazon/chronos-2) |
35
+ | TimesFM-2.5 | [google/timesfm-2.5-200m-pytorch](https://huggingface.co/google/timesfm-2.5-200m-pytorch) |
36
+ | FlowState | [ibm-research/flowstate](https://huggingface.co/ibm-research/flowstate) |
37
+ | TiReX | [NX-AI/TiRex-1.1-gifteval](https://huggingface.co/NX-AI/TiRex-1.1-gifteval) |
38
+ | PatchTST-FM | [ibm-research/patchtst-fm-r1](https://huggingface.co/ibm-research/patchtst-fm-r1) |
39
+ | Toto-2.0-2.5B | [Datadog/Toto-2.0-2.5B](https://huggingface.co/Datadog/Toto-2.0-2.5B) |
40
 
41
+ The original model repositories retain their respective licenses and terms.
 
 
 
 
 
 
 
42
 
43
+ ## Artifact Layout
44
 
45
+ ```text
46
+ CastStar-HF/
47
+ ├── README.md
48
+ ├── manifest.json
49
+ ├── calibration/
50
+ │ └── bucket_class_biases.csv
51
+ ├── models/
52
+ │ ├── pre/
53
+ │ └── post/
54
+ └── test/
55
+ ├── input_features/
56
+ ├── predictions/
57
+ └── predicted_features/
58
+ ```
59
 
60
+ The bundle contains only the frozen files required by the reproduction
61
+ notebook. It does not contain training data or training code.
62
 
63
+ ## Reproduction
64
 
65
+ 1. Clone and install [GIFT-Eval](https://github.com/SalesforceAIResearch/gift-eval).
66
+ 2. Download the [Salesforce/GiftEval](https://huggingface.co/datasets/Salesforce/GiftEval) dataset.
67
+ 3. Open [`notebooks/caststar.ipynb`](https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/caststar.ipynb).
68
+ 4. Set the local GIFT-Eval dataset path and run the notebook.
69
+
70
+ The notebook downloads this repository, runs frozen CastStar inference,
71
+ evaluates the forecasts with the official GIFT-Eval metrics, and writes a
72
+ submission-format CSV.
73
 
74
+ ## GIFT-Eval Submission
75
 
76
  ```json
77
  {
 
79
  "model_type": "agentic",
80
  "model_dtype": "float32",
81
  "model_link": "https://huggingface.co/USTC-AGI/CastStar",
82
+ "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/caststar.ipynb",
83
+ "org": "USTC-AGI",
84
  "testdata_leakage": "No",
85
+ "replication_code_available": "Yes"
86
  }
87
  ```
88
 
89
+ ## Intended Use
 
 
 
 
 
 
 
90
 
91
+ These assets are intended for reproducing and inspecting the CastStar
92
+ GIFT-Eval benchmark submission.
calibration/bucket_class_biases.csv ADDED
@@ -0,0 +1,121 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ bucket,model,bias
2
+ 10S|long,toto-2.0-2.5b,-0.8833333333333333
3
+ 10S|long,chronos-2,-1.1333333333333333
4
+ 10S|long,tirex,0.6666666666666666
5
+ 10S|long,patchtst-fm,1.0666666666666667
6
+ 10S|long,flowstate,-0.033333333333333326
7
+ 10S|long,timesfm-2.5,0.31666666666666665
8
+ 10S|medium,toto-2.0-2.5b,-0.4083333333333333
9
+ 10S|medium,chronos-2,-1.1583333333333332
10
+ 10S|medium,tirex,0.6416666666666666
11
+ 10S|medium,patchtst-fm,0.04166666666666663
12
+ 10S|medium,flowstate,0.5916666666666667
13
+ 10S|medium,timesfm-2.5,0.29166666666666663
14
+ 10S|short,toto-2.0-2.5b,-0.050000000000000044
15
+ 10S|short,chronos-2,-1.3
16
+ 10S|short,tirex,0.49999999999999994
17
+ 10S|short,patchtst-fm,-0.10000000000000003
18
+ 10S|short,flowstate,0.8
19
+ 10S|short,timesfm-2.5,0.14999999999999997
20
+ 10T|long,toto-2.0-2.5b,2.1416666666666666
21
+ 10T|long,chronos-2,0.46166666666666656
22
+ 10T|long,tirex,-3.4883333333333333
23
+ 10T|long,patchtst-fm,-0.33833333333333326
24
+ 10T|long,flowstate,2.0616666666666665
25
+ 10T|long,timesfm-2.5,-0.8383333333333333
26
+ 10T|medium,toto-2.0-2.5b,-1.1366666666666667
27
+ 10T|medium,chronos-2,0.8633333333333333
28
+ 10T|medium,tirex,-1.8366666666666667
29
+ 10T|medium,patchtst-fm,1.4833333333333334
30
+ 10T|medium,flowstate,0.8133333333333332
31
+ 10T|medium,timesfm-2.5,-0.18666666666666673
32
+ 10T|short,toto-2.0-2.5b,0.11666666666666664
33
+ 10T|short,chronos-2,-1.1333333333333333
34
+ 10T|short,tirex,0.6666666666666666
35
+ 10T|short,patchtst-fm,0.06666666666666665
36
+ 10T|short,flowstate,-0.033333333333333326
37
+ 10T|short,timesfm-2.5,0.31666666666666665
38
+ 15T|long,toto-2.0-2.5b,-0.31666666666666665
39
+ 15T|long,chronos-2,0.7833333333333334
40
+ 15T|long,tirex,0.9333333333333333
41
+ 15T|long,patchtst-fm,-0.5166666666666666
42
+ 15T|long,flowstate,-0.6166666666666667
43
+ 15T|long,timesfm-2.5,-0.26666666666666666
44
+ 15T|medium,toto-2.0-2.5b,0.6916666666666667
45
+ 15T|medium,chronos-2,-1.5383333333333333
46
+ 15T|medium,tirex,1.2616666666666667
47
+ 15T|medium,patchtst-fm,0.11166666666666665
48
+ 15T|medium,flowstate,-0.43833333333333335
49
+ 15T|medium,timesfm-2.5,-0.08833333333333336
50
+ 15T|short,toto-2.0-2.5b,-0.11333333333333329
51
+ 15T|short,chronos-2,0.3866666666666667
52
+ 15T|short,tirex,-0.31333333333333335
53
+ 15T|short,patchtst-fm,0.33666666666666667
54
+ 15T|short,flowstate,-0.38333333333333336
55
+ 15T|short,timesfm-2.5,0.08666666666666667
56
+ 5T|long,toto-2.0-2.5b,-0.62
57
+ 5T|long,chronos-2,-0.7000000000000001
58
+ 5T|long,tirex,-0.17000000000000015
59
+ 5T|long,patchtst-fm,-0.6200000000000002
60
+ 5T|long,flowstate,1.38
61
+ 5T|long,timesfm-2.5,0.7299999999999999
62
+ 5T|medium,toto-2.0-2.5b,-0.5216666666666667
63
+ 5T|medium,chronos-2,-2.0516666666666667
64
+ 5T|medium,tirex,-0.7016666666666667
65
+ 5T|medium,patchtst-fm,1.4783333333333333
66
+ 5T|medium,flowstate,1.3483333333333332
67
+ 5T|medium,timesfm-2.5,0.4483333333333332
68
+ 5T|short,toto-2.0-2.5b,0.11666666666666664
69
+ 5T|short,chronos-2,-1.1333333333333333
70
+ 5T|short,tirex,0.6666666666666666
71
+ 5T|short,patchtst-fm,0.06666666666666665
72
+ 5T|short,flowstate,-0.033333333333333326
73
+ 5T|short,timesfm-2.5,0.31666666666666665
74
+ A|short,toto-2.0-2.5b,-0.44499999999999995
75
+ A|short,chronos-2,-1.6750000000000003
76
+ A|short,tirex,-0.8749999999999999
77
+ A|short,patchtst-fm,0.795
78
+ A|short,flowstate,1.425
79
+ A|short,timesfm-2.5,0.775
80
+ D|short,toto-2.0-2.5b,0.2383333333333334
81
+ D|short,chronos-2,-0.5216666666666665
82
+ D|short,tirex,0.08833333333333337
83
+ D|short,patchtst-fm,-0.3316666666666666
84
+ D|short,flowstate,0.08833333333333343
85
+ D|short,timesfm-2.5,0.4383333333333334
86
+ H|long,toto-2.0-2.5b,-0.7166666666666666
87
+ H|long,chronos-2,-0.9666666666666666
88
+ H|long,tirex,0.8333333333333333
89
+ H|long,patchtst-fm,0.23333333333333334
90
+ H|long,flowstate,0.13333333333333336
91
+ H|long,timesfm-2.5,0.48333333333333334
92
+ H|medium,toto-2.0-2.5b,1.25
93
+ H|medium,chronos-2,-2.4
94
+ H|medium,tirex,1.2
95
+ H|medium,patchtst-fm,0.6
96
+ H|medium,flowstate,-1.25
97
+ H|medium,timesfm-2.5,0.6
98
+ H|short,toto-2.0-2.5b,1.0383333333333336
99
+ H|short,chronos-2,-1.7416666666666665
100
+ H|short,tirex,1.5283333333333335
101
+ H|short,patchtst-fm,-0.9916666666666665
102
+ H|short,flowstate,-1.0916666666666666
103
+ H|short,timesfm-2.5,1.2583333333333335
104
+ M|short,toto-2.0-2.5b,-0.9766666666666668
105
+ M|short,chronos-2,0.5033333333333334
106
+ M|short,tirex,-1.0966666666666667
107
+ M|short,patchtst-fm,0.21333333333333326
108
+ M|short,flowstate,0.3533333333333333
109
+ M|short,timesfm-2.5,1.0033333333333334
110
+ Q|short,toto-2.0-2.5b,0.53
111
+ Q|short,chronos-2,-1.32
112
+ Q|short,tirex,1.13
113
+ Q|short,patchtst-fm,-0.24999999999999997
114
+ Q|short,flowstate,-0.22
115
+ Q|short,timesfm-2.5,0.12999999999999998
116
+ W|short,toto-2.0-2.5b,0.43666666666666665
117
+ W|short,chronos-2,-1.7333333333333334
118
+ W|short,tirex,1.6666666666666665
119
+ W|short,patchtst-fm,-1.5333333333333332
120
+ W|short,flowstate,0.4466666666666667
121
+ W|short,timesfm-2.5,0.7166666666666667
manifest.json ADDED
@@ -0,0 +1,1652 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "format": "caststar-gift-eval-inference-v1",
3
+ "models": [
4
+ "toto-2.0-2.5b",
5
+ "chronos-2",
6
+ "tirex",
7
+ "patchtst-fm",
8
+ "flowstate",
9
+ "timesfm-2.5"
10
+ ],
11
+ "quantile_levels": [
12
+ 0.1,
13
+ 0.2,
14
+ 0.3,
15
+ 0.4,
16
+ 0.5,
17
+ 0.6,
18
+ 0.7,
19
+ 0.8,
20
+ 0.9
21
+ ],
22
+ "feature_columns": [
23
+ "hurst",
24
+ "series_length",
25
+ "unitroot_pp",
26
+ "unitroot_kpss",
27
+ "hw_alpha",
28
+ "hw_beta",
29
+ "hw_gamma",
30
+ "stability",
31
+ "nperiods",
32
+ "seasonal_period",
33
+ "trend",
34
+ "spike",
35
+ "linearity",
36
+ "curvature",
37
+ "e_acf1",
38
+ "e_acf10",
39
+ "seasonal_strength",
40
+ "peak",
41
+ "trough",
42
+ "x_pacf5",
43
+ "diff1x_pacf5",
44
+ "diff2x_pacf5",
45
+ "seas_pacf",
46
+ "nonlinearity",
47
+ "lumpiness",
48
+ "alpha",
49
+ "beta",
50
+ "arch_acf",
51
+ "garch_acf",
52
+ "arch_r2",
53
+ "garch_r2",
54
+ "flat_spots",
55
+ "entropy",
56
+ "crossing_points",
57
+ "arch_lm",
58
+ "x_acf1",
59
+ "x_acf10",
60
+ "diff1_acf1",
61
+ "diff1_acf10",
62
+ "diff2_acf1",
63
+ "diff2_acf10",
64
+ "seas_acf1",
65
+ "seasonality",
66
+ "prediction_length",
67
+ "num_variates",
68
+ "freq",
69
+ "domain"
70
+ ],
71
+ "categories": {
72
+ "domain": [
73
+ "Econ/Fin",
74
+ "Energy",
75
+ "Healthcare",
76
+ "Nature",
77
+ "Sales",
78
+ "Transport",
79
+ "Web/CloudOps"
80
+ ],
81
+ "freq": [
82
+ "10S",
83
+ "10T",
84
+ "15T",
85
+ "5T",
86
+ "A-DEC",
87
+ "D",
88
+ "H",
89
+ "M",
90
+ "Q-DEC",
91
+ "W-FRI",
92
+ "W-SUN",
93
+ "W-THU",
94
+ "W-TUE",
95
+ "W-WED"
96
+ ]
97
+ },
98
+ "configs": [
99
+ {
100
+ "dirname": "bitbrains_fast_storage_5T_long",
101
+ "dataset_name": "bitbrains_fast_storage/5T",
102
+ "dataset_config": "bitbrains_fast_storage/5T/long",
103
+ "bucket": "5T|long",
104
+ "freq": "5T",
105
+ "term": "long",
106
+ "prediction_length": 720,
107
+ "seasonality": 288,
108
+ "num_variates": 2,
109
+ "domain": "Web/CloudOps",
110
+ "windows": 2,
111
+ "test_rows": 5000,
112
+ "normalizer_mase": 1.1366422318753115,
113
+ "normalizer_crps": 1.1773077032683998
114
+ },
115
+ {
116
+ "dirname": "bitbrains_fast_storage_5T_medium",
117
+ "dataset_name": "bitbrains_fast_storage/5T",
118
+ "dataset_config": "bitbrains_fast_storage/5T/medium",
119
+ "bucket": "5T|medium",
120
+ "freq": "5T",
121
+ "term": "medium",
122
+ "prediction_length": 480,
123
+ "seasonality": 288,
124
+ "num_variates": 2,
125
+ "domain": "Web/CloudOps",
126
+ "windows": 2,
127
+ "test_rows": 5000,
128
+ "normalizer_mase": 1.2202713595882653,
129
+ "normalizer_crps": 1.1979128978088514
130
+ },
131
+ {
132
+ "dirname": "bitbrains_fast_storage_5T_short",
133
+ "dataset_name": "bitbrains_fast_storage/5T",
134
+ "dataset_config": "bitbrains_fast_storage/5T/short",
135
+ "bucket": "5T|short",
136
+ "freq": "5T",
137
+ "term": "short",
138
+ "prediction_length": 48,
139
+ "seasonality": 288,
140
+ "num_variates": 2,
141
+ "domain": "Web/CloudOps",
142
+ "windows": 18,
143
+ "test_rows": 45000,
144
+ "normalizer_mase": 1.1360241737534567,
145
+ "normalizer_crps": 1.210386425305063
146
+ },
147
+ {
148
+ "dirname": "bitbrains_fast_storage_H_short",
149
+ "dataset_name": "bitbrains_fast_storage/H",
150
+ "dataset_config": "bitbrains_fast_storage/H/short",
151
+ "bucket": "H|short",
152
+ "freq": "H",
153
+ "term": "short",
154
+ "prediction_length": 48,
155
+ "seasonality": 24,
156
+ "num_variates": 2,
157
+ "domain": "Web/CloudOps",
158
+ "windows": 2,
159
+ "test_rows": 5000,
160
+ "normalizer_mase": 1.2985124569638047,
161
+ "normalizer_crps": 1.0222395520114795
162
+ },
163
+ {
164
+ "dirname": "bitbrains_rnd_5T_long",
165
+ "dataset_name": "bitbrains_rnd/5T",
166
+ "dataset_config": "bitbrains_rnd/5T/long",
167
+ "bucket": "5T|long",
168
+ "freq": "5T",
169
+ "term": "long",
170
+ "prediction_length": 720,
171
+ "seasonality": 288,
172
+ "num_variates": 2,
173
+ "domain": "Web/CloudOps",
174
+ "windows": 2,
175
+ "test_rows": 2000,
176
+ "normalizer_mase": 3.501257904408163,
177
+ "normalizer_crps": 1.175220506770106
178
+ },
179
+ {
180
+ "dirname": "bitbrains_rnd_5T_medium",
181
+ "dataset_name": "bitbrains_rnd/5T",
182
+ "dataset_config": "bitbrains_rnd/5T/medium",
183
+ "bucket": "5T|medium",
184
+ "freq": "5T",
185
+ "term": "medium",
186
+ "prediction_length": 480,
187
+ "seasonality": 288,
188
+ "num_variates": 2,
189
+ "domain": "Web/CloudOps",
190
+ "windows": 2,
191
+ "test_rows": 2000,
192
+ "normalizer_mase": 4.542392314855541,
193
+ "normalizer_crps": 1.169291300089368
194
+ },
195
+ {
196
+ "dirname": "bitbrains_rnd_5T_short",
197
+ "dataset_name": "bitbrains_rnd/5T",
198
+ "dataset_config": "bitbrains_rnd/5T/short",
199
+ "bucket": "5T|short",
200
+ "freq": "5T",
201
+ "term": "short",
202
+ "prediction_length": 48,
203
+ "seasonality": 288,
204
+ "num_variates": 2,
205
+ "domain": "Web/CloudOps",
206
+ "windows": 18,
207
+ "test_rows": 18000,
208
+ "normalizer_mase": 1.9708126376716957,
209
+ "normalizer_crps": 1.1015904440824371
210
+ },
211
+ {
212
+ "dirname": "bitbrains_rnd_H_short",
213
+ "dataset_name": "bitbrains_rnd/H",
214
+ "dataset_config": "bitbrains_rnd/H/short",
215
+ "bucket": "H|short",
216
+ "freq": "H",
217
+ "term": "short",
218
+ "prediction_length": 48,
219
+ "seasonality": 24,
220
+ "num_variates": 2,
221
+ "domain": "Web/CloudOps",
222
+ "windows": 2,
223
+ "test_rows": 2000,
224
+ "normalizer_mase": 6.0373618687533925,
225
+ "normalizer_crps": 1.2432170953390527
226
+ },
227
+ {
228
+ "dirname": "bizitobs_application_10S_long",
229
+ "dataset_name": "bizitobs_application",
230
+ "dataset_config": "bizitobs_application/10S/long",
231
+ "bucket": "10S|long",
232
+ "freq": "10S",
233
+ "term": "long",
234
+ "prediction_length": 900,
235
+ "seasonality": 360,
236
+ "num_variates": 2,
237
+ "domain": "Web/CloudOps",
238
+ "windows": 1,
239
+ "test_rows": 2,
240
+ "normalizer_mase": 3.2063450561622804,
241
+ "normalizer_crps": 0.04573090228155046
242
+ },
243
+ {
244
+ "dirname": "bizitobs_application_10S_medium",
245
+ "dataset_name": "bizitobs_application",
246
+ "dataset_config": "bizitobs_application/10S/medium",
247
+ "bucket": "10S|medium",
248
+ "freq": "10S",
249
+ "term": "medium",
250
+ "prediction_length": 600,
251
+ "seasonality": 360,
252
+ "num_variates": 2,
253
+ "domain": "Web/CloudOps",
254
+ "windows": 2,
255
+ "test_rows": 4,
256
+ "normalizer_mase": 2.6914183385868085,
257
+ "normalizer_crps": 0.04268912394422428
258
+ },
259
+ {
260
+ "dirname": "bizitobs_application_10S_short",
261
+ "dataset_name": "bizitobs_application",
262
+ "dataset_config": "bizitobs_application/10S/short",
263
+ "bucket": "10S|short",
264
+ "freq": "10S",
265
+ "term": "short",
266
+ "prediction_length": 60,
267
+ "seasonality": 360,
268
+ "num_variates": 2,
269
+ "domain": "Web/CloudOps",
270
+ "windows": 15,
271
+ "test_rows": 30,
272
+ "normalizer_mase": 2.2423302386755415,
273
+ "normalizer_crps": 0.03483868247005456
274
+ },
275
+ {
276
+ "dirname": "bizitobs_l2c_5T_long",
277
+ "dataset_name": "bizitobs_l2c/5T",
278
+ "dataset_config": "bizitobs_l2c/5T/long",
279
+ "bucket": "5T|long",
280
+ "freq": "5T",
281
+ "term": "long",
282
+ "prediction_length": 720,
283
+ "seasonality": 288,
284
+ "num_variates": 7,
285
+ "domain": "Web/CloudOps",
286
+ "windows": 5,
287
+ "test_rows": 35,
288
+ "normalizer_mase": 1.4542830442631411,
289
+ "normalizer_crps": 0.6484921267253538
290
+ },
291
+ {
292
+ "dirname": "bizitobs_l2c_5T_medium",
293
+ "dataset_name": "bizitobs_l2c/5T",
294
+ "dataset_config": "bizitobs_l2c/5T/medium",
295
+ "bucket": "5T|medium",
296
+ "freq": "5T",
297
+ "term": "medium",
298
+ "prediction_length": 480,
299
+ "seasonality": 288,
300
+ "num_variates": 7,
301
+ "domain": "Web/CloudOps",
302
+ "windows": 7,
303
+ "test_rows": 49,
304
+ "normalizer_mase": 1.2435997989335994,
305
+ "normalizer_crps": 0.5203917750796179
306
+ },
307
+ {
308
+ "dirname": "bizitobs_l2c_5T_short",
309
+ "dataset_name": "bizitobs_l2c/5T",
310
+ "dataset_config": "bizitobs_l2c/5T/short",
311
+ "bucket": "5T|short",
312
+ "freq": "5T",
313
+ "term": "short",
314
+ "prediction_length": 48,
315
+ "seasonality": 288,
316
+ "num_variates": 7,
317
+ "domain": "Web/CloudOps",
318
+ "windows": 20,
319
+ "test_rows": 140,
320
+ "normalizer_mase": 0.9860210805885202,
321
+ "normalizer_crps": 0.26206838623738704
322
+ },
323
+ {
324
+ "dirname": "bizitobs_l2c_H_long",
325
+ "dataset_name": "bizitobs_l2c/H",
326
+ "dataset_config": "bizitobs_l2c/H/long",
327
+ "bucket": "H|long",
328
+ "freq": "H",
329
+ "term": "long",
330
+ "prediction_length": 720,
331
+ "seasonality": 24,
332
+ "num_variates": 7,
333
+ "domain": "Web/CloudOps",
334
+ "windows": 1,
335
+ "test_rows": 7,
336
+ "normalizer_mase": 1.4260541525613812,
337
+ "normalizer_crps": 0.9410651242377539
338
+ },
339
+ {
340
+ "dirname": "bizitobs_l2c_H_medium",
341
+ "dataset_name": "bizitobs_l2c/H",
342
+ "dataset_config": "bizitobs_l2c/H/medium",
343
+ "bucket": "H|medium",
344
+ "freq": "H",
345
+ "term": "medium",
346
+ "prediction_length": 480,
347
+ "seasonality": 24,
348
+ "num_variates": 7,
349
+ "domain": "Web/CloudOps",
350
+ "windows": 1,
351
+ "test_rows": 7,
352
+ "normalizer_mase": 1.5102861226255864,
353
+ "normalizer_crps": 0.9042051354742671
354
+ },
355
+ {
356
+ "dirname": "bizitobs_l2c_H_short",
357
+ "dataset_name": "bizitobs_l2c/H",
358
+ "dataset_config": "bizitobs_l2c/H/short",
359
+ "bucket": "H|short",
360
+ "freq": "H",
361
+ "term": "short",
362
+ "prediction_length": 48,
363
+ "seasonality": 24,
364
+ "num_variates": 7,
365
+ "domain": "Web/CloudOps",
366
+ "windows": 6,
367
+ "test_rows": 42,
368
+ "normalizer_mase": 1.2140641267600039,
369
+ "normalizer_crps": 0.5211675771895117
370
+ },
371
+ {
372
+ "dirname": "bizitobs_service_10S_long",
373
+ "dataset_name": "bizitobs_service",
374
+ "dataset_config": "bizitobs_service/10S/long",
375
+ "bucket": "10S|long",
376
+ "freq": "10S",
377
+ "term": "long",
378
+ "prediction_length": 900,
379
+ "seasonality": 360,
380
+ "num_variates": 2,
381
+ "domain": "Web/CloudOps",
382
+ "windows": 1,
383
+ "test_rows": 42,
384
+ "normalizer_mase": 1.367193469742303,
385
+ "normalizer_crps": 0.05345299584740217
386
+ },
387
+ {
388
+ "dirname": "bizitobs_service_10S_medium",
389
+ "dataset_name": "bizitobs_service",
390
+ "dataset_config": "bizitobs_service/10S/medium",
391
+ "bucket": "10S|medium",
392
+ "freq": "10S",
393
+ "term": "medium",
394
+ "prediction_length": 600,
395
+ "seasonality": 360,
396
+ "num_variates": 2,
397
+ "domain": "Web/CloudOps",
398
+ "windows": 2,
399
+ "test_rows": 84,
400
+ "normalizer_mase": 1.3205790727650968,
401
+ "normalizer_crps": 0.04755831292581585
402
+ },
403
+ {
404
+ "dirname": "bizitobs_service_10S_short",
405
+ "dataset_name": "bizitobs_service",
406
+ "dataset_config": "bizitobs_service/10S/short",
407
+ "bucket": "10S|short",
408
+ "freq": "10S",
409
+ "term": "short",
410
+ "prediction_length": 60,
411
+ "seasonality": 360,
412
+ "num_variates": 2,
413
+ "domain": "Web/CloudOps",
414
+ "windows": 15,
415
+ "test_rows": 630,
416
+ "normalizer_mase": 1.225305042620374,
417
+ "normalizer_crps": 0.03998233993351736
418
+ },
419
+ {
420
+ "dirname": "car_parts_M_short",
421
+ "dataset_name": "car_parts_with_missing",
422
+ "dataset_config": "car_parts/M/short",
423
+ "bucket": "M|short",
424
+ "freq": "M",
425
+ "term": "short",
426
+ "prediction_length": 12,
427
+ "seasonality": 12,
428
+ "num_variates": 1,
429
+ "domain": "Sales",
430
+ "windows": 1,
431
+ "test_rows": 2674,
432
+ "normalizer_mase": 1.2014638390969912,
433
+ "normalizer_crps": 1.7217438941460104
434
+ },
435
+ {
436
+ "dirname": "covid_deaths_D_short",
437
+ "dataset_name": "covid_deaths",
438
+ "dataset_config": "covid_deaths/D/short",
439
+ "bucket": "D|short",
440
+ "freq": "D",
441
+ "term": "short",
442
+ "prediction_length": 30,
443
+ "seasonality": 1,
444
+ "num_variates": 1,
445
+ "domain": "Healthcare",
446
+ "windows": 1,
447
+ "test_rows": 266,
448
+ "normalizer_mase": 46.91239825526407,
449
+ "normalizer_crps": 0.12672327831223676
450
+ },
451
+ {
452
+ "dirname": "electricity_15T_long",
453
+ "dataset_name": "electricity/15T",
454
+ "dataset_config": "electricity/15T/long",
455
+ "bucket": "15T|long",
456
+ "freq": "15T",
457
+ "term": "long",
458
+ "prediction_length": 720,
459
+ "seasonality": 96,
460
+ "num_variates": 1,
461
+ "domain": "Energy",
462
+ "windows": 20,
463
+ "test_rows": 7400,
464
+ "normalizer_mase": 1.1635317075319347,
465
+ "normalizer_crps": 0.11255332426055248
466
+ },
467
+ {
468
+ "dirname": "electricity_15T_medium",
469
+ "dataset_name": "electricity/15T",
470
+ "dataset_config": "electricity/15T/medium",
471
+ "bucket": "15T|medium",
472
+ "freq": "15T",
473
+ "term": "medium",
474
+ "prediction_length": 480,
475
+ "seasonality": 96,
476
+ "num_variates": 1,
477
+ "domain": "Energy",
478
+ "windows": 20,
479
+ "test_rows": 7400,
480
+ "normalizer_mase": 1.1507882041044137,
481
+ "normalizer_crps": 0.11275485280084273
482
+ },
483
+ {
484
+ "dirname": "electricity_15T_short",
485
+ "dataset_name": "electricity/15T",
486
+ "dataset_config": "electricity/15T/short",
487
+ "bucket": "15T|short",
488
+ "freq": "15T",
489
+ "term": "short",
490
+ "prediction_length": 48,
491
+ "seasonality": 96,
492
+ "num_variates": 1,
493
+ "domain": "Energy",
494
+ "windows": 20,
495
+ "test_rows": 7400,
496
+ "normalizer_mase": 1.7170601412372184,
497
+ "normalizer_crps": 0.16489261500071847
498
+ },
499
+ {
500
+ "dirname": "electricity_D_short",
501
+ "dataset_name": "electricity/D",
502
+ "dataset_config": "electricity/D/short",
503
+ "bucket": "D|short",
504
+ "freq": "D",
505
+ "term": "short",
506
+ "prediction_length": 30,
507
+ "seasonality": 1,
508
+ "num_variates": 1,
509
+ "domain": "Energy",
510
+ "windows": 5,
511
+ "test_rows": 1850,
512
+ "normalizer_mase": 1.987041399842847,
513
+ "normalizer_crps": 0.10405987098536745
514
+ },
515
+ {
516
+ "dirname": "electricity_H_long",
517
+ "dataset_name": "electricity/H",
518
+ "dataset_config": "electricity/H/long",
519
+ "bucket": "H|long",
520
+ "freq": "H",
521
+ "term": "long",
522
+ "prediction_length": 720,
523
+ "seasonality": 24,
524
+ "num_variates": 1,
525
+ "domain": "Energy",
526
+ "windows": 5,
527
+ "test_rows": 1850,
528
+ "normalizer_mase": 1.5240786163496851,
529
+ "normalizer_crps": 0.15341839756227935
530
+ },
531
+ {
532
+ "dirname": "electricity_H_medium",
533
+ "dataset_name": "electricity/H",
534
+ "dataset_config": "electricity/H/medium",
535
+ "bucket": "H|medium",
536
+ "freq": "H",
537
+ "term": "medium",
538
+ "prediction_length": 480,
539
+ "seasonality": 24,
540
+ "num_variates": 1,
541
+ "domain": "Energy",
542
+ "windows": 8,
543
+ "test_rows": 2960,
544
+ "normalizer_mase": 1.3924537851702354,
545
+ "normalizer_crps": 0.12738701179892217
546
+ },
547
+ {
548
+ "dirname": "electricity_H_short",
549
+ "dataset_name": "electricity/H",
550
+ "dataset_config": "electricity/H/short",
551
+ "bucket": "H|short",
552
+ "freq": "H",
553
+ "term": "short",
554
+ "prediction_length": 48,
555
+ "seasonality": 24,
556
+ "num_variates": 1,
557
+ "domain": "Energy",
558
+ "windows": 20,
559
+ "test_rows": 7400,
560
+ "normalizer_mase": 1.3577726051465433,
561
+ "normalizer_crps": 0.10566499518474065
562
+ },
563
+ {
564
+ "dirname": "electricity_W_short",
565
+ "dataset_name": "electricity/W",
566
+ "dataset_config": "electricity/W-FRI/short",
567
+ "bucket": "W|short",
568
+ "freq": "W-FRI",
569
+ "term": "short",
570
+ "prediction_length": 8,
571
+ "seasonality": 1,
572
+ "num_variates": 1,
573
+ "domain": "Energy",
574
+ "windows": 3,
575
+ "test_rows": 1110,
576
+ "normalizer_mase": 2.0896995064573947,
577
+ "normalizer_crps": 0.0993032024444414
578
+ },
579
+ {
580
+ "dirname": "ett1_15T_long",
581
+ "dataset_name": "ett1/15T",
582
+ "dataset_config": "ett1/15T/long",
583
+ "bucket": "15T|long",
584
+ "freq": "15T",
585
+ "term": "long",
586
+ "prediction_length": 720,
587
+ "seasonality": 96,
588
+ "num_variates": 7,
589
+ "domain": "Energy",
590
+ "windows": 10,
591
+ "test_rows": 70,
592
+ "normalizer_mase": 1.190885705867336,
593
+ "normalizer_crps": 0.3401793468576481
594
+ },
595
+ {
596
+ "dirname": "ett1_15T_medium",
597
+ "dataset_name": "ett1/15T",
598
+ "dataset_config": "ett1/15T/medium",
599
+ "bucket": "15T|medium",
600
+ "freq": "15T",
601
+ "term": "medium",
602
+ "prediction_length": 480,
603
+ "seasonality": 96,
604
+ "num_variates": 7,
605
+ "domain": "Energy",
606
+ "windows": 15,
607
+ "test_rows": 105,
608
+ "normalizer_mase": 1.188305978388322,
609
+ "normalizer_crps": 0.32155116949786394
610
+ },
611
+ {
612
+ "dirname": "ett1_15T_short",
613
+ "dataset_name": "ett1/15T",
614
+ "dataset_config": "ett1/15T/short",
615
+ "bucket": "15T|short",
616
+ "freq": "15T",
617
+ "term": "short",
618
+ "prediction_length": 48,
619
+ "seasonality": 96,
620
+ "num_variates": 7,
621
+ "domain": "Energy",
622
+ "windows": 20,
623
+ "test_rows": 140,
624
+ "normalizer_mase": 0.9341549587802876,
625
+ "normalizer_crps": 0.24136394938567346
626
+ },
627
+ {
628
+ "dirname": "ett1_D_short",
629
+ "dataset_name": "ett1/D",
630
+ "dataset_config": "ett1/D/short",
631
+ "bucket": "D|short",
632
+ "freq": "D",
633
+ "term": "short",
634
+ "prediction_length": 30,
635
+ "seasonality": 1,
636
+ "num_variates": 7,
637
+ "domain": "Energy",
638
+ "windows": 3,
639
+ "test_rows": 21,
640
+ "normalizer_mase": 1.7783697877928102,
641
+ "normalizer_crps": 0.40843357791503654
642
+ },
643
+ {
644
+ "dirname": "ett1_H_long",
645
+ "dataset_name": "ett1/H",
646
+ "dataset_config": "ett1/H/long",
647
+ "bucket": "H|long",
648
+ "freq": "H",
649
+ "term": "long",
650
+ "prediction_length": 720,
651
+ "seasonality": 24,
652
+ "num_variates": 7,
653
+ "domain": "Energy",
654
+ "windows": 3,
655
+ "test_rows": 21,
656
+ "normalizer_mase": 1.4787012870437326,
657
+ "normalizer_crps": 0.47106248989542887
658
+ },
659
+ {
660
+ "dirname": "ett1_H_medium",
661
+ "dataset_name": "ett1/H",
662
+ "dataset_config": "ett1/H/medium",
663
+ "bucket": "H|medium",
664
+ "freq": "H",
665
+ "term": "medium",
666
+ "prediction_length": 480,
667
+ "seasonality": 24,
668
+ "num_variates": 7,
669
+ "domain": "Energy",
670
+ "windows": 4,
671
+ "test_rows": 28,
672
+ "normalizer_mase": 1.5678007461285395,
673
+ "normalizer_crps": 0.4349118298516912
674
+ },
675
+ {
676
+ "dirname": "ett1_H_short",
677
+ "dataset_name": "ett1/H",
678
+ "dataset_config": "ett1/H/short",
679
+ "bucket": "H|short",
680
+ "freq": "H",
681
+ "term": "short",
682
+ "prediction_length": 48,
683
+ "seasonality": 24,
684
+ "num_variates": 7,
685
+ "domain": "Energy",
686
+ "windows": 20,
687
+ "test_rows": 140,
688
+ "normalizer_mase": 0.9773176243434534,
689
+ "normalizer_crps": 0.24036844107833774
690
+ },
691
+ {
692
+ "dirname": "ett1_W_short",
693
+ "dataset_name": "ett1/W",
694
+ "dataset_config": "ett1/W-THU/short",
695
+ "bucket": "W|short",
696
+ "freq": "W-THU",
697
+ "term": "short",
698
+ "prediction_length": 8,
699
+ "seasonality": 1,
700
+ "num_variates": 7,
701
+ "domain": "Energy",
702
+ "windows": 2,
703
+ "test_rows": 14,
704
+ "normalizer_mase": 1.7689207877683926,
705
+ "normalizer_crps": 0.3118251468740881
706
+ },
707
+ {
708
+ "dirname": "ett2_15T_long",
709
+ "dataset_name": "ett2/15T",
710
+ "dataset_config": "ett2/15T/long",
711
+ "bucket": "15T|long",
712
+ "freq": "15T",
713
+ "term": "long",
714
+ "prediction_length": 720,
715
+ "seasonality": 96,
716
+ "num_variates": 7,
717
+ "domain": "Energy",
718
+ "windows": 10,
719
+ "test_rows": 70,
720
+ "normalizer_mase": 1.0126577751121104,
721
+ "normalizer_crps": 0.13282444765695306
722
+ },
723
+ {
724
+ "dirname": "ett2_15T_medium",
725
+ "dataset_name": "ett2/15T",
726
+ "dataset_config": "ett2/15T/medium",
727
+ "bucket": "15T|medium",
728
+ "freq": "15T",
729
+ "term": "medium",
730
+ "prediction_length": 480,
731
+ "seasonality": 96,
732
+ "num_variates": 7,
733
+ "domain": "Energy",
734
+ "windows": 15,
735
+ "test_rows": 105,
736
+ "normalizer_mase": 1.0512391501712754,
737
+ "normalizer_crps": 0.12412877798161984
738
+ },
739
+ {
740
+ "dirname": "ett2_15T_short",
741
+ "dataset_name": "ett2/15T",
742
+ "dataset_config": "ett2/15T/short",
743
+ "bucket": "15T|short",
744
+ "freq": "15T",
745
+ "term": "short",
746
+ "prediction_length": 48,
747
+ "seasonality": 96,
748
+ "num_variates": 7,
749
+ "domain": "Energy",
750
+ "windows": 20,
751
+ "test_rows": 140,
752
+ "normalizer_mase": 1.067314800011479,
753
+ "normalizer_crps": 0.09638438775638664
754
+ },
755
+ {
756
+ "dirname": "ett2_D_short",
757
+ "dataset_name": "ett2/D",
758
+ "dataset_config": "ett2/D/short",
759
+ "bucket": "D|short",
760
+ "freq": "D",
761
+ "term": "short",
762
+ "prediction_length": 30,
763
+ "seasonality": 1,
764
+ "num_variates": 7,
765
+ "domain": "Energy",
766
+ "windows": 3,
767
+ "test_rows": 21,
768
+ "normalizer_mase": 1.3901143157307532,
769
+ "normalizer_crps": 0.15331100213050436
770
+ },
771
+ {
772
+ "dirname": "ett2_H_long",
773
+ "dataset_name": "ett2/H",
774
+ "dataset_config": "ett2/H/long",
775
+ "bucket": "H|long",
776
+ "freq": "H",
777
+ "term": "long",
778
+ "prediction_length": 720,
779
+ "seasonality": 24,
780
+ "num_variates": 7,
781
+ "domain": "Energy",
782
+ "windows": 3,
783
+ "test_rows": 21,
784
+ "normalizer_mase": 1.1284016697207502,
785
+ "normalizer_crps": 0.20790633225768584
786
+ },
787
+ {
788
+ "dirname": "ett2_H_medium",
789
+ "dataset_name": "ett2/H",
790
+ "dataset_config": "ett2/H/medium",
791
+ "bucket": "H|medium",
792
+ "freq": "H",
793
+ "term": "medium",
794
+ "prediction_length": 480,
795
+ "seasonality": 24,
796
+ "num_variates": 7,
797
+ "domain": "Energy",
798
+ "windows": 4,
799
+ "test_rows": 28,
800
+ "normalizer_mase": 1.2384249730532626,
801
+ "normalizer_crps": 0.1862050814081793
802
+ },
803
+ {
804
+ "dirname": "ett2_H_short",
805
+ "dataset_name": "ett2/H",
806
+ "dataset_config": "ett2/H/short",
807
+ "bucket": "H|short",
808
+ "freq": "H",
809
+ "term": "short",
810
+ "prediction_length": 48,
811
+ "seasonality": 24,
812
+ "num_variates": 7,
813
+ "domain": "Energy",
814
+ "windows": 20,
815
+ "test_rows": 140,
816
+ "normalizer_mase": 0.9231678277377344,
817
+ "normalizer_crps": 0.08895289185095483
818
+ },
819
+ {
820
+ "dirname": "ett2_W_short",
821
+ "dataset_name": "ett2/W",
822
+ "dataset_config": "ett2/W-THU/short",
823
+ "bucket": "W|short",
824
+ "freq": "W-THU",
825
+ "term": "short",
826
+ "prediction_length": 8,
827
+ "seasonality": 1,
828
+ "num_variates": 7,
829
+ "domain": "Energy",
830
+ "windows": 2,
831
+ "test_rows": 14,
832
+ "normalizer_mase": 0.7785209533197451,
833
+ "normalizer_crps": 0.13367758870767996
834
+ },
835
+ {
836
+ "dirname": "hierarchical_sales_D_short",
837
+ "dataset_name": "hierarchical_sales/D",
838
+ "dataset_config": "hierarchical_sales/D/short",
839
+ "bucket": "D|short",
840
+ "freq": "D",
841
+ "term": "short",
842
+ "prediction_length": 30,
843
+ "seasonality": 1,
844
+ "num_variates": 1,
845
+ "domain": "Sales",
846
+ "windows": 7,
847
+ "test_rows": 826,
848
+ "normalizer_mase": 1.1348033996861053,
849
+ "normalizer_crps": 1.7364572065613715
850
+ },
851
+ {
852
+ "dirname": "hierarchical_sales_W_short",
853
+ "dataset_name": "hierarchical_sales/W",
854
+ "dataset_config": "hierarchical_sales/W-WED/short",
855
+ "bucket": "W|short",
856
+ "freq": "W-WED",
857
+ "term": "short",
858
+ "prediction_length": 8,
859
+ "seasonality": 1,
860
+ "num_variates": 1,
861
+ "domain": "Sales",
862
+ "windows": 4,
863
+ "test_rows": 472,
864
+ "normalizer_mase": 1.0250133777288337,
865
+ "normalizer_crps": 0.8322267526416594
866
+ },
867
+ {
868
+ "dirname": "hospital_M_short",
869
+ "dataset_name": "hospital",
870
+ "dataset_config": "hospital/M/short",
871
+ "bucket": "M|short",
872
+ "freq": "M",
873
+ "term": "short",
874
+ "prediction_length": 12,
875
+ "seasonality": 12,
876
+ "num_variates": 1,
877
+ "domain": "Healthcare",
878
+ "windows": 1,
879
+ "test_rows": 767,
880
+ "normalizer_mase": 0.9205278266364826,
881
+ "normalizer_crps": 0.06248817651596496
882
+ },
883
+ {
884
+ "dirname": "jena_weather_10T_long",
885
+ "dataset_name": "jena_weather/10T",
886
+ "dataset_config": "jena_weather/10T/long",
887
+ "bucket": "10T|long",
888
+ "freq": "10T",
889
+ "term": "long",
890
+ "prediction_length": 720,
891
+ "seasonality": 144,
892
+ "num_variates": 21,
893
+ "domain": "Nature",
894
+ "windows": 8,
895
+ "test_rows": 168,
896
+ "normalizer_mase": 0.7614923703210493,
897
+ "normalizer_crps": 0.2371949131202996
898
+ },
899
+ {
900
+ "dirname": "jena_weather_10T_medium",
901
+ "dataset_name": "jena_weather/10T",
902
+ "dataset_config": "jena_weather/10T/medium",
903
+ "bucket": "10T|medium",
904
+ "freq": "10T",
905
+ "term": "medium",
906
+ "prediction_length": 480,
907
+ "seasonality": 144,
908
+ "num_variates": 21,
909
+ "domain": "Nature",
910
+ "windows": 11,
911
+ "test_rows": 231,
912
+ "normalizer_mase": 0.7160100391723763,
913
+ "normalizer_crps": 0.21183609008512827
914
+ },
915
+ {
916
+ "dirname": "jena_weather_10T_short",
917
+ "dataset_name": "jena_weather/10T",
918
+ "dataset_config": "jena_weather/10T/short",
919
+ "bucket": "10T|short",
920
+ "freq": "10T",
921
+ "term": "short",
922
+ "prediction_length": 48,
923
+ "seasonality": 144,
924
+ "num_variates": 21,
925
+ "domain": "Nature",
926
+ "windows": 20,
927
+ "test_rows": 420,
928
+ "normalizer_mase": 0.7428874675800381,
929
+ "normalizer_crps": 0.15520092375681044
930
+ },
931
+ {
932
+ "dirname": "jena_weather_D_short",
933
+ "dataset_name": "jena_weather/D",
934
+ "dataset_config": "jena_weather/D/short",
935
+ "bucket": "D|short",
936
+ "freq": "D",
937
+ "term": "short",
938
+ "prediction_length": 30,
939
+ "seasonality": 1,
940
+ "num_variates": 21,
941
+ "domain": "Nature",
942
+ "windows": 2,
943
+ "test_rows": 42,
944
+ "normalizer_mase": 1.5734140297468562,
945
+ "normalizer_crps": 0.2105908566035542
946
+ },
947
+ {
948
+ "dirname": "jena_weather_H_long",
949
+ "dataset_name": "jena_weather/H",
950
+ "dataset_config": "jena_weather/H/long",
951
+ "bucket": "H|long",
952
+ "freq": "H",
953
+ "term": "long",
954
+ "prediction_length": 720,
955
+ "seasonality": 24,
956
+ "num_variates": 21,
957
+ "domain": "Nature",
958
+ "windows": 2,
959
+ "test_rows": 42,
960
+ "normalizer_mase": 1.2677178473920059,
961
+ "normalizer_crps": 0.41914340130808575
962
+ },
963
+ {
964
+ "dirname": "jena_weather_H_medium",
965
+ "dataset_name": "jena_weather/H",
966
+ "dataset_config": "jena_weather/H/medium",
967
+ "bucket": "H|medium",
968
+ "freq": "H",
969
+ "term": "medium",
970
+ "prediction_length": 480,
971
+ "seasonality": 24,
972
+ "num_variates": 21,
973
+ "domain": "Nature",
974
+ "windows": 2,
975
+ "test_rows": 42,
976
+ "normalizer_mase": 0.8886133410507664,
977
+ "normalizer_crps": 0.3428055217231077
978
+ },
979
+ {
980
+ "dirname": "jena_weather_H_short",
981
+ "dataset_name": "jena_weather/H",
982
+ "dataset_config": "jena_weather/H/short",
983
+ "bucket": "H|short",
984
+ "freq": "H",
985
+ "term": "short",
986
+ "prediction_length": 48,
987
+ "seasonality": 24,
988
+ "num_variates": 21,
989
+ "domain": "Nature",
990
+ "windows": 19,
991
+ "test_rows": 399,
992
+ "normalizer_mase": 0.7229102185557323,
993
+ "normalizer_crps": 0.15420788787522166
994
+ },
995
+ {
996
+ "dirname": "kdd_cup_2018_D_short",
997
+ "dataset_name": "kdd_cup_2018_with_missing/D",
998
+ "dataset_config": "kdd_cup_2018/D/short",
999
+ "bucket": "D|short",
1000
+ "freq": "D",
1001
+ "term": "short",
1002
+ "prediction_length": 30,
1003
+ "seasonality": 1,
1004
+ "num_variates": 1,
1005
+ "domain": "Nature",
1006
+ "windows": 2,
1007
+ "test_rows": 540,
1008
+ "normalizer_mase": 1.4970343738270926,
1009
+ "normalizer_crps": 0.674506129145155
1010
+ },
1011
+ {
1012
+ "dirname": "kdd_cup_2018_H_long",
1013
+ "dataset_name": "kdd_cup_2018_with_missing/H",
1014
+ "dataset_config": "kdd_cup_2018/H/long",
1015
+ "bucket": "H|long",
1016
+ "freq": "H",
1017
+ "term": "long",
1018
+ "prediction_length": 720,
1019
+ "seasonality": 24,
1020
+ "num_variates": 1,
1021
+ "domain": "Nature",
1022
+ "windows": 2,
1023
+ "test_rows": 540,
1024
+ "normalizer_mase": 1.3355336658089667,
1025
+ "normalizer_crps": 0.9362833347539716
1026
+ },
1027
+ {
1028
+ "dirname": "kdd_cup_2018_H_medium",
1029
+ "dataset_name": "kdd_cup_2018_with_missing/H",
1030
+ "dataset_config": "kdd_cup_2018/H/medium",
1031
+ "bucket": "H|medium",
1032
+ "freq": "H",
1033
+ "term": "medium",
1034
+ "prediction_length": 480,
1035
+ "seasonality": 24,
1036
+ "num_variates": 1,
1037
+ "domain": "Nature",
1038
+ "windows": 2,
1039
+ "test_rows": 540,
1040
+ "normalizer_mase": 1.4289997072062277,
1041
+ "normalizer_crps": 0.7587892026846513
1042
+ },
1043
+ {
1044
+ "dirname": "kdd_cup_2018_H_short",
1045
+ "dataset_name": "kdd_cup_2018_with_missing/H",
1046
+ "dataset_config": "kdd_cup_2018/H/short",
1047
+ "bucket": "H|short",
1048
+ "freq": "H",
1049
+ "term": "short",
1050
+ "prediction_length": 48,
1051
+ "seasonality": 24,
1052
+ "num_variates": 1,
1053
+ "domain": "Nature",
1054
+ "windows": 20,
1055
+ "test_rows": 5400,
1056
+ "normalizer_mase": 1.340425278558367,
1057
+ "normalizer_crps": 0.547536580422314
1058
+ },
1059
+ {
1060
+ "dirname": "loop_seattle_5T_long",
1061
+ "dataset_name": "LOOP_SEATTLE/5T",
1062
+ "dataset_config": "loop_seattle/5T/long",
1063
+ "bucket": "5T|long",
1064
+ "freq": "5T",
1065
+ "term": "long",
1066
+ "prediction_length": 720,
1067
+ "seasonality": 288,
1068
+ "num_variates": 1,
1069
+ "domain": "Transport",
1070
+ "windows": 15,
1071
+ "test_rows": 4845,
1072
+ "normalizer_mase": 1.2506857542807588,
1073
+ "normalizer_crps": 0.12716656087996273
1074
+ },
1075
+ {
1076
+ "dirname": "loop_seattle_5T_medium",
1077
+ "dataset_name": "LOOP_SEATTLE/5T",
1078
+ "dataset_config": "loop_seattle/5T/medium",
1079
+ "bucket": "5T|medium",
1080
+ "freq": "5T",
1081
+ "term": "medium",
1082
+ "prediction_length": 480,
1083
+ "seasonality": 288,
1084
+ "num_variates": 1,
1085
+ "domain": "Transport",
1086
+ "windows": 20,
1087
+ "test_rows": 6460,
1088
+ "normalizer_mase": 1.1532851467900647,
1089
+ "normalizer_crps": 0.11727937998585615
1090
+ },
1091
+ {
1092
+ "dirname": "loop_seattle_5T_short",
1093
+ "dataset_name": "LOOP_SEATTLE/5T",
1094
+ "dataset_config": "loop_seattle/5T/short",
1095
+ "bucket": "5T|short",
1096
+ "freq": "5T",
1097
+ "term": "short",
1098
+ "prediction_length": 48,
1099
+ "seasonality": 288,
1100
+ "num_variates": 1,
1101
+ "domain": "Transport",
1102
+ "windows": 20,
1103
+ "test_rows": 6460,
1104
+ "normalizer_mase": 0.7623397118337274,
1105
+ "normalizer_crps": 0.08083202568463041
1106
+ },
1107
+ {
1108
+ "dirname": "loop_seattle_D_short",
1109
+ "dataset_name": "LOOP_SEATTLE/D",
1110
+ "dataset_config": "loop_seattle/D/short",
1111
+ "bucket": "D|short",
1112
+ "freq": "D",
1113
+ "term": "short",
1114
+ "prediction_length": 30,
1115
+ "seasonality": 1,
1116
+ "num_variates": 1,
1117
+ "domain": "Transport",
1118
+ "windows": 2,
1119
+ "test_rows": 646,
1120
+ "normalizer_mase": 1.7323957056549062,
1121
+ "normalizer_crps": 0.1032547596878987
1122
+ },
1123
+ {
1124
+ "dirname": "loop_seattle_H_long",
1125
+ "dataset_name": "LOOP_SEATTLE/H",
1126
+ "dataset_config": "loop_seattle/H/long",
1127
+ "bucket": "H|long",
1128
+ "freq": "H",
1129
+ "term": "long",
1130
+ "prediction_length": 720,
1131
+ "seasonality": 24,
1132
+ "num_variates": 1,
1133
+ "domain": "Transport",
1134
+ "windows": 2,
1135
+ "test_rows": 646,
1136
+ "normalizer_mase": 1.5460268753677606,
1137
+ "normalizer_crps": 0.18709450197084374
1138
+ },
1139
+ {
1140
+ "dirname": "loop_seattle_H_medium",
1141
+ "dataset_name": "LOOP_SEATTLE/H",
1142
+ "dataset_config": "loop_seattle/H/medium",
1143
+ "bucket": "H|medium",
1144
+ "freq": "H",
1145
+ "term": "medium",
1146
+ "prediction_length": 480,
1147
+ "seasonality": 24,
1148
+ "num_variates": 1,
1149
+ "domain": "Transport",
1150
+ "windows": 2,
1151
+ "test_rows": 646,
1152
+ "normalizer_mase": 1.4805126192799751,
1153
+ "normalizer_crps": 0.16208090743299983
1154
+ },
1155
+ {
1156
+ "dirname": "loop_seattle_H_short",
1157
+ "dataset_name": "LOOP_SEATTLE/H",
1158
+ "dataset_config": "loop_seattle/H/short",
1159
+ "bucket": "H|short",
1160
+ "freq": "H",
1161
+ "term": "short",
1162
+ "prediction_length": 48,
1163
+ "seasonality": 24,
1164
+ "num_variates": 1,
1165
+ "domain": "Transport",
1166
+ "windows": 19,
1167
+ "test_rows": 6137,
1168
+ "normalizer_mase": 1.2928412613913995,
1169
+ "normalizer_crps": 0.10423508438727377
1170
+ },
1171
+ {
1172
+ "dirname": "m4_daily_D_short",
1173
+ "dataset_name": "m4_daily",
1174
+ "dataset_config": "m4_daily/D/short",
1175
+ "bucket": "D|short",
1176
+ "freq": "D",
1177
+ "term": "short",
1178
+ "prediction_length": 14,
1179
+ "seasonality": 1,
1180
+ "num_variates": 1,
1181
+ "domain": "Econ/Fin",
1182
+ "windows": 1,
1183
+ "test_rows": 4227,
1184
+ "normalizer_mase": 3.278424296610481,
1185
+ "normalizer_crps": 0.024356147648337902
1186
+ },
1187
+ {
1188
+ "dirname": "m4_hourly_H_short",
1189
+ "dataset_name": "m4_hourly",
1190
+ "dataset_config": "m4_hourly/H/short",
1191
+ "bucket": "H|short",
1192
+ "freq": "H",
1193
+ "term": "short",
1194
+ "prediction_length": 48,
1195
+ "seasonality": 24,
1196
+ "num_variates": 1,
1197
+ "domain": "Econ/Fin",
1198
+ "windows": 1,
1199
+ "test_rows": 414,
1200
+ "normalizer_mase": 1.1932101877276544,
1201
+ "normalizer_crps": 0.03757255147212195
1202
+ },
1203
+ {
1204
+ "dirname": "m4_monthly_M_short",
1205
+ "dataset_name": "m4_monthly",
1206
+ "dataset_config": "m4_monthly/M/short",
1207
+ "bucket": "M|short",
1208
+ "freq": "M",
1209
+ "term": "short",
1210
+ "prediction_length": 18,
1211
+ "seasonality": 12,
1212
+ "num_variates": 1,
1213
+ "domain": "Econ/Fin",
1214
+ "windows": 1,
1215
+ "test_rows": 48000,
1216
+ "normalizer_mase": 1.259717038974429,
1217
+ "normalizer_crps": 0.12192063664829357
1218
+ },
1219
+ {
1220
+ "dirname": "m4_quarterly_Q_short",
1221
+ "dataset_name": "m4_quarterly",
1222
+ "dataset_config": "m4_quarterly/Q-DEC/short",
1223
+ "bucket": "Q|short",
1224
+ "freq": "Q-DEC",
1225
+ "term": "short",
1226
+ "prediction_length": 8,
1227
+ "seasonality": 4,
1228
+ "num_variates": 1,
1229
+ "domain": "Econ/Fin",
1230
+ "windows": 1,
1231
+ "test_rows": 24000,
1232
+ "normalizer_mase": 1.602247174616002,
1233
+ "normalizer_crps": 0.09805397548466407
1234
+ },
1235
+ {
1236
+ "dirname": "m4_weekly_W_short",
1237
+ "dataset_name": "m4_weekly",
1238
+ "dataset_config": "m4_weekly/W-SUN/short",
1239
+ "bucket": "W|short",
1240
+ "freq": "W-SUN",
1241
+ "term": "short",
1242
+ "prediction_length": 13,
1243
+ "seasonality": 1,
1244
+ "num_variates": 1,
1245
+ "domain": "Econ/Fin",
1246
+ "windows": 1,
1247
+ "test_rows": 359,
1248
+ "normalizer_mase": 2.777295047362158,
1249
+ "normalizer_crps": 0.06087039452311706
1250
+ },
1251
+ {
1252
+ "dirname": "m4_yearly_A_short",
1253
+ "dataset_name": "m4_yearly",
1254
+ "dataset_config": "m4_yearly/A-DEC/short",
1255
+ "bucket": "A|short",
1256
+ "freq": "A-DEC",
1257
+ "term": "short",
1258
+ "prediction_length": 6,
1259
+ "seasonality": 1,
1260
+ "num_variates": 1,
1261
+ "domain": "Econ/Fin",
1262
+ "windows": 1,
1263
+ "test_rows": 22974,
1264
+ "normalizer_mase": 3.965954931783283,
1265
+ "normalizer_crps": 0.13714565740983498
1266
+ },
1267
+ {
1268
+ "dirname": "m_dense_D_short",
1269
+ "dataset_name": "M_DENSE/D",
1270
+ "dataset_config": "m_dense/D/short",
1271
+ "bucket": "D|short",
1272
+ "freq": "D",
1273
+ "term": "short",
1274
+ "prediction_length": 30,
1275
+ "seasonality": 1,
1276
+ "num_variates": 1,
1277
+ "domain": "Transport",
1278
+ "windows": 3,
1279
+ "test_rows": 90,
1280
+ "normalizer_mase": 1.6693368812171987,
1281
+ "normalizer_crps": 0.22686565739428607
1282
+ },
1283
+ {
1284
+ "dirname": "m_dense_H_long",
1285
+ "dataset_name": "M_DENSE/H",
1286
+ "dataset_config": "m_dense/H/long",
1287
+ "bucket": "H|long",
1288
+ "freq": "H",
1289
+ "term": "long",
1290
+ "prediction_length": 720,
1291
+ "seasonality": 24,
1292
+ "num_variates": 1,
1293
+ "domain": "Transport",
1294
+ "windows": 3,
1295
+ "test_rows": 90,
1296
+ "normalizer_mase": 1.4779178744488217,
1297
+ "normalizer_crps": 0.4189507904607682
1298
+ },
1299
+ {
1300
+ "dirname": "m_dense_H_medium",
1301
+ "dataset_name": "M_DENSE/H",
1302
+ "dataset_config": "m_dense/H/medium",
1303
+ "bucket": "H|medium",
1304
+ "freq": "H",
1305
+ "term": "medium",
1306
+ "prediction_length": 480,
1307
+ "seasonality": 24,
1308
+ "num_variates": 1,
1309
+ "domain": "Transport",
1310
+ "windows": 4,
1311
+ "test_rows": 120,
1312
+ "normalizer_mase": 1.569946511898904,
1313
+ "normalizer_crps": 0.3773835994986822
1314
+ },
1315
+ {
1316
+ "dirname": "m_dense_H_short",
1317
+ "dataset_name": "M_DENSE/H",
1318
+ "dataset_config": "m_dense/H/short",
1319
+ "bucket": "H|short",
1320
+ "freq": "H",
1321
+ "term": "short",
1322
+ "prediction_length": 48,
1323
+ "seasonality": 24,
1324
+ "num_variates": 1,
1325
+ "domain": "Transport",
1326
+ "windows": 20,
1327
+ "test_rows": 600,
1328
+ "normalizer_mase": 1.4875444172240992,
1329
+ "normalizer_crps": 0.27479786933969136
1330
+ },
1331
+ {
1332
+ "dirname": "restaurant_D_short",
1333
+ "dataset_name": "restaurant",
1334
+ "dataset_config": "restaurant/D/short",
1335
+ "bucket": "D|short",
1336
+ "freq": "D",
1337
+ "term": "short",
1338
+ "prediction_length": 30,
1339
+ "seasonality": 1,
1340
+ "num_variates": 1,
1341
+ "domain": "Sales",
1342
+ "windows": 1,
1343
+ "test_rows": 807,
1344
+ "normalizer_mase": 1.0060757450046092,
1345
+ "normalizer_crps": 0.6770164746085788
1346
+ },
1347
+ {
1348
+ "dirname": "saugeen_D_short",
1349
+ "dataset_name": "saugeenday/D",
1350
+ "dataset_config": "saugeen/D/short",
1351
+ "bucket": "D|short",
1352
+ "freq": "D",
1353
+ "term": "short",
1354
+ "prediction_length": 30,
1355
+ "seasonality": 1,
1356
+ "num_variates": 1,
1357
+ "domain": "Nature",
1358
+ "windows": 20,
1359
+ "test_rows": 20,
1360
+ "normalizer_mase": 3.413049305768806,
1361
+ "normalizer_crps": 0.5850141736335629
1362
+ },
1363
+ {
1364
+ "dirname": "saugeen_M_short",
1365
+ "dataset_name": "saugeenday/M",
1366
+ "dataset_config": "saugeen/M/short",
1367
+ "bucket": "M|short",
1368
+ "freq": "M",
1369
+ "term": "short",
1370
+ "prediction_length": 12,
1371
+ "seasonality": 12,
1372
+ "num_variates": 1,
1373
+ "domain": "Nature",
1374
+ "windows": 7,
1375
+ "test_rows": 7,
1376
+ "normalizer_mase": 0.9763741730697157,
1377
+ "normalizer_crps": 0.4450846286544009
1378
+ },
1379
+ {
1380
+ "dirname": "saugeen_W_short",
1381
+ "dataset_name": "saugeenday/W",
1382
+ "dataset_config": "saugeen/W-THU/short",
1383
+ "bucket": "W|short",
1384
+ "freq": "W-THU",
1385
+ "term": "short",
1386
+ "prediction_length": 8,
1387
+ "seasonality": 1,
1388
+ "num_variates": 1,
1389
+ "domain": "Nature",
1390
+ "windows": 20,
1391
+ "test_rows": 20,
1392
+ "normalizer_mase": 1.990658004828632,
1393
+ "normalizer_crps": 0.7340352210434381
1394
+ },
1395
+ {
1396
+ "dirname": "solar_10T_long",
1397
+ "dataset_name": "solar/10T",
1398
+ "dataset_config": "solar/10T/long",
1399
+ "bucket": "10T|long",
1400
+ "freq": "10T",
1401
+ "term": "long",
1402
+ "prediction_length": 720,
1403
+ "seasonality": 144,
1404
+ "num_variates": 1,
1405
+ "domain": "Energy",
1406
+ "windows": 8,
1407
+ "test_rows": 1096,
1408
+ "normalizer_mase": 0.8709317964102903,
1409
+ "normalizer_crps": 0.6738494837655387
1410
+ },
1411
+ {
1412
+ "dirname": "solar_10T_medium",
1413
+ "dataset_name": "solar/10T",
1414
+ "dataset_config": "solar/10T/medium",
1415
+ "bucket": "10T|medium",
1416
+ "freq": "10T",
1417
+ "term": "medium",
1418
+ "prediction_length": 480,
1419
+ "seasonality": 144,
1420
+ "num_variates": 1,
1421
+ "domain": "Energy",
1422
+ "windows": 11,
1423
+ "test_rows": 1507,
1424
+ "normalizer_mase": 0.9270005404761702,
1425
+ "normalizer_crps": 0.6550635536405177
1426
+ },
1427
+ {
1428
+ "dirname": "solar_10T_short",
1429
+ "dataset_name": "solar/10T",
1430
+ "dataset_config": "solar/10T/short",
1431
+ "bucket": "10T|short",
1432
+ "freq": "10T",
1433
+ "term": "short",
1434
+ "prediction_length": 48,
1435
+ "seasonality": 144,
1436
+ "num_variates": 1,
1437
+ "domain": "Energy",
1438
+ "windows": 20,
1439
+ "test_rows": 2740,
1440
+ "normalizer_mase": 1.1057028063509833,
1441
+ "normalizer_crps": 0.8595386998005952
1442
+ },
1443
+ {
1444
+ "dirname": "solar_D_short",
1445
+ "dataset_name": "solar/D",
1446
+ "dataset_config": "solar/D/short",
1447
+ "bucket": "D|short",
1448
+ "freq": "D",
1449
+ "term": "short",
1450
+ "prediction_length": 30,
1451
+ "seasonality": 1,
1452
+ "num_variates": 1,
1453
+ "domain": "Energy",
1454
+ "windows": 2,
1455
+ "test_rows": 274,
1456
+ "normalizer_mase": 1.155861190610263,
1457
+ "normalizer_crps": 0.5590651164917229
1458
+ },
1459
+ {
1460
+ "dirname": "solar_H_long",
1461
+ "dataset_name": "solar/H",
1462
+ "dataset_config": "solar/H/long",
1463
+ "bucket": "H|long",
1464
+ "freq": "H",
1465
+ "term": "long",
1466
+ "prediction_length": 720,
1467
+ "seasonality": 24,
1468
+ "num_variates": 1,
1469
+ "domain": "Energy",
1470
+ "windows": 2,
1471
+ "test_rows": 274,
1472
+ "normalizer_mase": 1.0711966477927781,
1473
+ "normalizer_crps": 1.0779764182077312
1474
+ },
1475
+ {
1476
+ "dirname": "solar_H_medium",
1477
+ "dataset_name": "solar/H",
1478
+ "dataset_config": "solar/H/medium",
1479
+ "bucket": "H|medium",
1480
+ "freq": "H",
1481
+ "term": "medium",
1482
+ "prediction_length": 480,
1483
+ "seasonality": 24,
1484
+ "num_variates": 1,
1485
+ "domain": "Energy",
1486
+ "windows": 2,
1487
+ "test_rows": 274,
1488
+ "normalizer_mase": 0.9349685908749716,
1489
+ "normalizer_crps": 0.9459874672135314
1490
+ },
1491
+ {
1492
+ "dirname": "solar_H_short",
1493
+ "dataset_name": "solar/H",
1494
+ "dataset_config": "solar/H/short",
1495
+ "bucket": "H|short",
1496
+ "freq": "H",
1497
+ "term": "short",
1498
+ "prediction_length": 48,
1499
+ "seasonality": 24,
1500
+ "num_variates": 1,
1501
+ "domain": "Energy",
1502
+ "windows": 19,
1503
+ "test_rows": 2603,
1504
+ "normalizer_mase": 0.9519362458316395,
1505
+ "normalizer_crps": 0.5916559631947232
1506
+ },
1507
+ {
1508
+ "dirname": "solar_W_short",
1509
+ "dataset_name": "solar/W",
1510
+ "dataset_config": "solar/W-FRI/short",
1511
+ "bucket": "W|short",
1512
+ "freq": "W-FRI",
1513
+ "term": "short",
1514
+ "prediction_length": 8,
1515
+ "seasonality": 1,
1516
+ "num_variates": 1,
1517
+ "domain": "Energy",
1518
+ "windows": 1,
1519
+ "test_rows": 137,
1520
+ "normalizer_mase": 1.4703506784640206,
1521
+ "normalizer_crps": 0.20974762813186765
1522
+ },
1523
+ {
1524
+ "dirname": "sz_taxi_15T_long",
1525
+ "dataset_name": "SZ_TAXI/15T",
1526
+ "dataset_config": "sz_taxi/15T/long",
1527
+ "bucket": "15T|long",
1528
+ "freq": "15T",
1529
+ "term": "long",
1530
+ "prediction_length": 720,
1531
+ "seasonality": 96,
1532
+ "num_variates": 1,
1533
+ "domain": "Transport",
1534
+ "windows": 1,
1535
+ "test_rows": 156,
1536
+ "normalizer_mase": 0.6909431344941248,
1537
+ "normalizer_crps": 0.4279704063043717
1538
+ },
1539
+ {
1540
+ "dirname": "sz_taxi_15T_medium",
1541
+ "dataset_name": "SZ_TAXI/15T",
1542
+ "dataset_config": "sz_taxi/15T/medium",
1543
+ "bucket": "15T|medium",
1544
+ "freq": "15T",
1545
+ "term": "medium",
1546
+ "prediction_length": 480,
1547
+ "seasonality": 96,
1548
+ "num_variates": 1,
1549
+ "domain": "Transport",
1550
+ "windows": 1,
1551
+ "test_rows": 156,
1552
+ "normalizer_mase": 0.7134840540400068,
1553
+ "normalizer_crps": 0.379172158612487
1554
+ },
1555
+ {
1556
+ "dirname": "sz_taxi_15T_short",
1557
+ "dataset_name": "SZ_TAXI/15T",
1558
+ "dataset_config": "sz_taxi/15T/short",
1559
+ "bucket": "15T|short",
1560
+ "freq": "15T",
1561
+ "term": "short",
1562
+ "prediction_length": 48,
1563
+ "seasonality": 96,
1564
+ "num_variates": 1,
1565
+ "domain": "Transport",
1566
+ "windows": 7,
1567
+ "test_rows": 1092,
1568
+ "normalizer_mase": 0.7644167735543712,
1569
+ "normalizer_crps": 0.3087068505119876
1570
+ },
1571
+ {
1572
+ "dirname": "sz_taxi_H_short",
1573
+ "dataset_name": "SZ_TAXI/H",
1574
+ "dataset_config": "sz_taxi/H/short",
1575
+ "bucket": "H|short",
1576
+ "freq": "H",
1577
+ "term": "short",
1578
+ "prediction_length": 48,
1579
+ "seasonality": 24,
1580
+ "num_variates": 1,
1581
+ "domain": "Transport",
1582
+ "windows": 2,
1583
+ "test_rows": 312,
1584
+ "normalizer_mase": 0.7381672994945998,
1585
+ "normalizer_crps": 0.21382494112013678
1586
+ },
1587
+ {
1588
+ "dirname": "temperature_rain_D_short",
1589
+ "dataset_name": "temperature_rain_with_missing",
1590
+ "dataset_config": "temperature_rain/D/short",
1591
+ "bucket": "D|short",
1592
+ "freq": "D",
1593
+ "term": "short",
1594
+ "prediction_length": 30,
1595
+ "seasonality": 1,
1596
+ "num_variates": 1,
1597
+ "domain": "Nature",
1598
+ "windows": 3,
1599
+ "test_rows": 96216,
1600
+ "normalizer_mase": 2.0115357653222166,
1601
+ "normalizer_crps": 1.2679455200160903
1602
+ },
1603
+ {
1604
+ "dirname": "us_births_D_short",
1605
+ "dataset_name": "us_births/D",
1606
+ "dataset_config": "us_births/D/short",
1607
+ "bucket": "D|short",
1608
+ "freq": "D",
1609
+ "term": "short",
1610
+ "prediction_length": 30,
1611
+ "seasonality": 1,
1612
+ "num_variates": 1,
1613
+ "domain": "Healthcare",
1614
+ "windows": 20,
1615
+ "test_rows": 20,
1616
+ "normalizer_mase": 1.8648391912491094,
1617
+ "normalizer_crps": 0.11952686561898423
1618
+ },
1619
+ {
1620
+ "dirname": "us_births_M_short",
1621
+ "dataset_name": "us_births/M",
1622
+ "dataset_config": "us_births/M/short",
1623
+ "bucket": "M|short",
1624
+ "freq": "M",
1625
+ "term": "short",
1626
+ "prediction_length": 12,
1627
+ "seasonality": 12,
1628
+ "num_variates": 1,
1629
+ "domain": "Healthcare",
1630
+ "windows": 2,
1631
+ "test_rows": 2,
1632
+ "normalizer_mase": 0.7605310792868888,
1633
+ "normalizer_crps": 0.016825596814792485
1634
+ },
1635
+ {
1636
+ "dirname": "us_births_W_short",
1637
+ "dataset_name": "us_births/W",
1638
+ "dataset_config": "us_births/W-TUE/short",
1639
+ "bucket": "W|short",
1640
+ "freq": "W-TUE",
1641
+ "term": "short",
1642
+ "prediction_length": 8,
1643
+ "seasonality": 1,
1644
+ "num_variates": 1,
1645
+ "domain": "Healthcare",
1646
+ "windows": 14,
1647
+ "test_rows": 14,
1648
+ "normalizer_mase": 1.5634204337929731,
1649
+ "normalizer_crps": 0.01929679613291714
1650
+ }
1651
+ ]
1652
+ }
models/pre/models.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ [
2
+ "toto-2.0-2.5b",
3
+ "chronos-2",
4
+ "tirex",
5
+ "patchtst-fm",
6
+ "flowstate",
7
+ "timesfm-2.5"
8
+ ]
test/predicted_features/flowstate/bitbrains_fast_storage_5T_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d1d42b295e0611d831af5d4fbea484e5ed6634b4f4c413545eff6e2138939409
3
+ size 564878
test/predicted_features/flowstate/bitbrains_fast_storage_5T_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:213aaa10be4f85a990344d9a0081a89d4258c7f82a13d93ae7a95ad670252b05
3
+ size 414686
test/predicted_features/flowstate/bitbrains_fast_storage_5T_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c09a9827a3b04e342aa0aed4762697f96ecaf0e5fca76695eb0b935c0595d9b0
3
+ size 3494243
test/predicted_features/flowstate/bitbrains_fast_storage_H_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4d5b6782315e37dd2476bea210bb554d58bb88b0aa0906eed493ed552cd0857b
3
+ size 479823
test/predicted_features/flowstate/bitbrains_rnd_5T_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:adc0769b9d44131534287c69923b140e353485b47bb33551893210a43472dc28
3
+ size 223119
test/predicted_features/flowstate/bitbrains_rnd_5T_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7496c8fe2f3cfe05a5bb06b4b326a6665b7b8aa2bbe0da2ffca5e4dbbcf01cf
3
+ size 163661
test/predicted_features/flowstate/bitbrains_rnd_5T_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:440f9dc4f0d0aa2163df0d66cf8cc4648f8cf583237375371ee00e8042168413
3
+ size 1375096
test/predicted_features/flowstate/bitbrains_rnd_H_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4aa66f3347ec67de7a132748a9caccccf1becaa26b0db206568db77a838e7074
3
+ size 190741
test/predicted_features/flowstate/bizitobs_application_10S_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a364e2d25b3af89a84415bb0d453fdb5d8ee316321e5908a68acc96d632efa45
3
+ size 2685
test/predicted_features/flowstate/bizitobs_application_10S_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6cbdad9c69cbad5134381cde2d4cf2071ca7021cc063af18437f83e96f93d9c2
3
+ size 2836
test/predicted_features/flowstate/bizitobs_application_10S_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9476506b588dd932d554e4f888780b08eb80f28711864e045a7485c82856876
3
+ size 5200
test/predicted_features/flowstate/bizitobs_l2c_5T_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1cb95642df2b4759ea52cf452afaa7d73cf7cd8916388581ecf504a476bc62de
3
+ size 6941
test/predicted_features/flowstate/bizitobs_l2c_5T_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00da63f4b78371df9c922f0cfddd0a7314ea95d57ef1b44d2a63a85584bbfb29
3
+ size 6960
test/predicted_features/flowstate/bizitobs_l2c_5T_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f50453eed05754f32761c19779d7af8137fd0d37665d44869171d59e15ac415e
3
+ size 14718
test/predicted_features/flowstate/bizitobs_l2c_H_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:decb574c7d964568dc1f1664c90deaf6ac32f30e7bac45177456afcca234a4f6
3
+ size 3367
test/predicted_features/flowstate/bizitobs_l2c_H_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:290e343cff1b27cd2c72c8f551f5d755b362544d91f169f99f2c6a525d39ff9a
3
+ size 3375
test/predicted_features/flowstate/bizitobs_l2c_H_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd94c5e2f5d3c24563dff676706dec772cb00ec1cc3a69554ffc1735a3c54361
3
+ size 7196
test/predicted_features/flowstate/bizitobs_service_10S_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d04ed7be8f11c4a5ae036303000d959a80025940f92e12bf48cd5490b5be49b9
3
+ size 7869
test/predicted_features/flowstate/bizitobs_service_10S_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cf3950ea922d219624b9b222c2a980764df881e1a432b0443a059509db5c302a
3
+ size 10448
test/predicted_features/flowstate/bizitobs_service_10S_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9c4c5f7ede0ced07af1d30212d7c0eaaf83a4ca993445e4b44fa7d421a7f84f
3
+ size 57741
test/predicted_features/flowstate/car_parts_M_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ac3e234952e5ce8df7722cdd738c41048a8fd72c69cbdecc7fe38da6828447bf
3
+ size 198097
test/predicted_features/flowstate/covid_deaths_D_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f58914e29928510fa34f14d7e193ebd525fb7509cfb75e05179435d19e2bc1a
3
+ size 20867
test/predicted_features/flowstate/electricity_15T_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b85e43c02ee81047858679278c3c24e64d1fb208c96535cb5fd43ab8400c9fe
3
+ size 877253
test/predicted_features/flowstate/electricity_15T_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3f084de2a7452108d61b83026c78cb049f8e732edb31dbb8019fe8e212d15ce
3
+ size 868037
test/predicted_features/flowstate/electricity_15T_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba498c617f657c7f1e3c2b44fefd984538d9d31fc496433782155d30b5d31895
3
+ size 604502
test/predicted_features/flowstate/electricity_D_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d7bfff39879b6e865802fba38482761842bfd17783e34cd66d7eec79bffd9bbb
3
+ size 146651
test/predicted_features/flowstate/electricity_H_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ca935b1af994f6f61375c4fc00f6fc378b90d4c1484b0b4a368e734b9c42379
3
+ size 227156
test/predicted_features/timesfm-2.5/loop_seattle_H_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b8935d0fcbcd5bbacb47a1bb13b8bd51a3807f1c8e9eb7b046e3601fee26300d
3
+ size 633455
test/predicted_features/timesfm-2.5/m4_quarterly_Q_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:79d61b924dda35a9d6b6e2f187895abacb3cf4b3767588894752da78cbf22c62
3
+ size 1580210
test/predicted_features/timesfm-2.5/saugeen_D_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:971285954df5c512d04018374d263bb9feabe04498516f0c3f8087883db722a5
3
+ size 4155
test/predicted_features/timesfm-2.5/saugeen_W_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f885c2823ad3ab7ed0c0b768ae4d48a698d5b0302fb5be63c9edc6ac6c53b504
3
+ size 3216
test/predicted_features/timesfm-2.5/solar_10T_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d30efed856287f1f39fa412a4dd8a68244628a4e1fbf0c55128f4a46fd054ea
3
+ size 135006
test/predicted_features/timesfm-2.5/solar_10T_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb77a35fa665fa3a51ea826b9b2f26c6b99e9a59aaa5461ba242663ba1cd5422
3
+ size 184021
test/predicted_features/timesfm-2.5/solar_10T_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71ffb160f201ee344bb46cdfbd5588dca582361ce7a1fce42cc911a087664991
3
+ size 225870
test/predicted_features/timesfm-2.5/solar_D_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:27dfadeb9c833ca6d3a644124407676bb988c3f732b9b89681be023c42e3cc79
3
+ size 24487
test/predicted_features/timesfm-2.5/solar_H_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9342d3258c7ef3c8111172d6fa294c7db70b695030af543e5812b84dcecee87d
3
+ size 33071
test/predicted_features/timesfm-2.5/solar_H_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb01f4aa5bcc286316b146e7fabb997fd8a4f8fdb812608ac437bab8413177e1
3
+ size 33170
test/predicted_features/timesfm-2.5/solar_H_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:898f219554156467232bdbd681edd7077f45f26f6d024e3666e481f07ba1da09
3
+ size 254497
test/predicted_features/timesfm-2.5/solar_W_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a14e423fe3755142971d91cd1b8177bc4fe9bf7a3c113abeca95d5776856451b
3
+ size 7110
test/predicted_features/timesfm-2.5/sz_taxi_15T_long/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d15fcd10b2211aba5fb95b663a134dd0b164f63317f04f5b59feeb46ec3f741
3
+ size 22440
test/predicted_features/timesfm-2.5/sz_taxi_15T_medium/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d765db6bdf727999439e179dcfc543dbe1c3fc2563a78a77685770fe92d04e33
3
+ size 22204
test/predicted_features/timesfm-2.5/sz_taxi_15T_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:045e6c141c255a122ac16878814af6b3792fcf9ff721e30ef5444330a4b82422
3
+ size 96873
test/predicted_features/timesfm-2.5/sz_taxi_H_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:009dc1468d4fd4f7dfb3b1be0f017208bd6c4231506c3aae720e6a4b2e5b4a93
3
+ size 35494
test/predicted_features/timesfm-2.5/us_births_D_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:893caf097f825b0f7eb4481bfdc93f133948c79ad40f549ed18024933e04289d
3
+ size 4111
test/predicted_features/timesfm-2.5/us_births_M_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eaa86e2bb23313ddf9db7a3ee005d47441b68a1e3f9aadadbd505ab0bdc708fa
3
+ size 2582
test/predicted_features/timesfm-2.5/us_births_W_short/predicted_window_features.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d67f35523c4174f4ac3139209a5e341c95031e1918ce7477a3aa54fb2803bffd
3
+ size 2961