Taha Aksu commited on
Commit
8c443bb
·
1 Parent(s): da0a655

add open source information

Browse files
Files changed (47) hide show
  1. app.py +7 -5
  2. results/Chronos_small/config.json +2 -1
  3. results/DLinear/config.json +2 -1
  4. results/FlowState-9.1M/config.json +2 -1
  5. results/Lag-Llama/config.json +2 -1
  6. results/Moirai2/config.json +2 -1
  7. results/Moirai_base/config.json +2 -1
  8. results/Moirai_large/config.json +2 -1
  9. results/Moirai_small/config.json +2 -1
  10. results/N-BEATS/config.json +2 -1
  11. results/PatchTST/config.json +2 -1
  12. results/TSOrchestra-test/config.json +3 -2
  13. results/TSOrchestra/config.json +3 -2
  14. results/TTM-R1-Pretrained/config.json +2 -1
  15. results/TTM-R2-Finetuned/config.json +3 -2
  16. results/TTM-R2-Pretrained/config.json +2 -1
  17. results/TiRex/config.json +2 -1
  18. results/TimeCopilot/config.json +2 -1
  19. results/TimesFM-2.5/config.json +3 -2
  20. results/Toto_Open_Base_1.0/config.json +2 -1
  21. results/YingLong_110m/config.json +2 -1
  22. results/YingLong_300m/config.json +2 -1
  23. results/YingLong_50m/config.json +2 -1
  24. results/YingLong_6m/config.json +2 -1
  25. results/auto_arima/config.json +2 -1
  26. results/auto_ets/config.json +2 -1
  27. results/auto_theta/config.json +2 -1
  28. results/chronos_base/config.json +2 -1
  29. results/chronos_bolt_base/config.json +2 -1
  30. results/chronos_bolt_small/config.json +2 -1
  31. results/chronos_large/config.json +2 -1
  32. results/crossformer/config.json +2 -1
  33. results/deepar/config.json +2 -1
  34. results/granite-flowstate-r1/config.json +2 -1
  35. results/iTransformer/config.json +2 -1
  36. results/naive/config.json +2 -1
  37. results/seasonal_naive/config.json +2 -1
  38. results/sundial_base_128m/config.json +2 -1
  39. results/tabpfn_ts/config.json +2 -1
  40. results/tempo_ensemble/config.json +2 -1
  41. results/tft/config.json +2 -1
  42. results/tide/config.json +2 -1
  43. results/timesfm/config.json +2 -1
  44. results/timesfm_2_0_500m/config.json +2 -1
  45. results/visionts/config.json +2 -1
  46. src/display/utils.py +2 -1
  47. src/leaderboard/read_evals.py +4 -1
app.py CHANGED
@@ -121,7 +121,7 @@ def init_leaderboard(ori_dataframe, model_info_df, sort_val: str | list | None =
121
  merged_df = merged_df[new_cols]
122
  if sort_val:
123
  if isinstance(sort_val, list):
124
- assert sort_val[0] == 'TestData Leakage'
125
  # ipdb.set_trace()
126
  leakage_order = pd.Categorical(merged_df[sort_val[0]], categories=['No', 'Yes', 'N/A'], ordered=True)
127
  merged_df['leakage_order'] = leakage_order
@@ -170,10 +170,11 @@ def init_leaderboard(ori_dataframe, model_info_df, sort_val: str | list | None =
170
  # ],
171
  filter_columns=[
172
  ColumnFilter(ModelInfoColumn.model_type.name, type="checkboxgroup", label="Model types"),
173
- ColumnFilter(ModelInfoColumn.testdata_leakage.name, type="checkboxgroup", label="TestData Leakage"),
 
174
  ],
175
  # bool_checkboxgroup_label="",
176
- column_widths=[40, 150] + [180 for _ in range(len(merged_df.columns)-2)],
177
  interactive=False,
178
  )
179
 
@@ -185,7 +186,7 @@ with demo:
185
  with gr.Tabs(elem_classes="tab-buttons") as tabs:
186
  with gr.TabItem('🏅 Overall', elem_id="llm-benchmark-tab-table", id=5):
187
  # leaderboard = init_leaderboard(overall_df, model_info_df, sort_val='Rank')
188
- leaderboard = init_leaderboard(overall_df, model_info_df, sort_val=['TestData Leakage', 'MASE_Rank'])
189
  print(f'FINAL Overall LEADERBOARD {overall_df}')
190
  with gr.TabItem("🏅 By Domain", elem_id="llm-benchmark-tab-table", id=0):
191
  leaderboard = init_leaderboard(domain_df, model_info_df)
@@ -217,7 +218,7 @@ with demo:
217
 
218
  if (!target) { return []; } // safety guard
219
 
220
- // Ask Gradios front-end to re-compute its filters:
221
  target.dispatchEvent(new Event('input', { bubbles: true }));
222
 
223
  return []; // load() must return something
@@ -225,6 +226,7 @@ with demo:
225
  """
226
  )
227
 
 
228
  with gr.Row():
229
  with gr.Accordion("📙 Citation", open=False):
230
  citation_button = gr.Textbox(
 
121
  merged_df = merged_df[new_cols]
122
  if sort_val:
123
  if isinstance(sort_val, list):
124
+ assert sort_val[0] == 'Test Leak.'
125
  # ipdb.set_trace()
126
  leakage_order = pd.Categorical(merged_df[sort_val[0]], categories=['No', 'Yes', 'N/A'], ordered=True)
127
  merged_df['leakage_order'] = leakage_order
 
170
  # ],
171
  filter_columns=[
172
  ColumnFilter(ModelInfoColumn.model_type.name, type="checkboxgroup", label="Model types"),
173
+ ColumnFilter(ModelInfoColumn.testdata_leakage.name, type="checkboxgroup", label="Test Leak."),
174
+ ColumnFilter(ModelInfoColumn.code_available.name, type="checkboxgroup", label="OSS"),
175
  ],
176
  # bool_checkboxgroup_label="",
177
+ column_widths=[30, 180] + [130 for _ in range(len(merged_df.columns)-2)],
178
  interactive=False,
179
  )
180
 
 
186
  with gr.Tabs(elem_classes="tab-buttons") as tabs:
187
  with gr.TabItem('🏅 Overall', elem_id="llm-benchmark-tab-table", id=5):
188
  # leaderboard = init_leaderboard(overall_df, model_info_df, sort_val='Rank')
189
+ leaderboard = init_leaderboard(overall_df, model_info_df, sort_val=['Test Leak.', 'MASE_Rank'])
190
  print(f'FINAL Overall LEADERBOARD {overall_df}')
191
  with gr.TabItem("🏅 By Domain", elem_id="llm-benchmark-tab-table", id=0):
192
  leaderboard = init_leaderboard(domain_df, model_info_df)
 
218
 
219
  if (!target) { return []; } // safety guard
220
 
221
+ // Ask Gradio's front-end to re-compute its filters:
222
  target.dispatchEvent(new Event('input', { bubbles: true }));
223
 
224
  return []; // load() must return something
 
226
  """
227
  )
228
 
229
+
230
  with gr.Row():
231
  with gr.Accordion("📙 Citation", open=False):
232
  citation_button = gr.Textbox(
results/Chronos_small/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/amazon/chronos-t5-small",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
- "testdata_leakage": "Yes"
 
9
  }
 
5
  "model_link": "https://huggingface.co/amazon/chronos-t5-small",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
  }
results/DLinear/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "The Chinese University of Hong Kong",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "The Chinese University of Hong Kong",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/FlowState-9.1M/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/ibm-research/flowstate",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/flowstate.ipynb",
7
  "org": "IBM Research",
8
- "testdata_leakage": "No"
 
9
  }
 
5
  "model_link": "https://huggingface.co/ibm-research/flowstate",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/flowstate.ipynb",
7
  "org": "IBM Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
  }
results/Lag-Llama/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/time-series-foundation-models/Lag-Llama",
6
  "org": "Morgan Stanley & Service Now",
7
- "testdata_leakage": "Yes"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/time-series-foundation-models/Lag-Llama",
6
  "org": "Morgan Stanley & Service Now",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "Yes"
9
  }
results/Moirai2/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/Salesforce/moirai-2.0-R-small",
6
  "code_link": "https://github.com/SalesforceAIResearch/uni2ts",
7
  "org": "Salesforce AI Research",
8
- "testdata_leakage": "No"
 
9
  }
 
5
  "model_link": "https://huggingface.co/Salesforce/moirai-2.0-R-small",
6
  "code_link": "https://github.com/SalesforceAIResearch/uni2ts",
7
  "org": "Salesforce AI Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
  }
results/Moirai_base/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/Salesforce/moirai-1.1-R-base",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/moirai.ipynb",
7
  "org": "Salesforce AI Research",
8
- "testdata_leakage": "No"
 
9
  }
 
5
  "model_link": "https://huggingface.co/Salesforce/moirai-1.1-R-base",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/moirai.ipynb",
7
  "org": "Salesforce AI Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
  }
results/Moirai_large/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/Salesforce/moirai-1.1-R-large",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/moirai.ipynb",
7
  "org": "Salesforce AI Research",
8
- "testdata_leakage": "No"
 
9
  }
 
5
  "model_link": "https://huggingface.co/Salesforce/moirai-1.1-R-large",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/moirai.ipynb",
7
  "org": "Salesforce AI Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
  }
results/Moirai_small/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/Salesforce/moirai-1.1-R-large",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/moirai.ipynb",
7
  "org": "Salesforce AI Research",
8
- "testdata_leakage": "No"
 
9
  }
 
5
  "model_link": "https://huggingface.co/Salesforce/moirai-1.1-R-large",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/moirai.ipynb",
7
  "org": "Salesforce AI Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
  }
results/N-BEATS/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "ServiceNow",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "ServiceNow",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/PatchTST/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Princeton University",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Princeton University",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/TSOrchestra-test/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/Melady/TEMPO",
6
  "org": "Melady Lab @ USC",
7
- "testdata_leakage": "Yes"
8
- }
 
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/Melady/TEMPO",
6
  "org": "Melady Lab @ USC",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "No"
9
+ }
results/TSOrchestra/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/DC-research/TSorchestra",
6
  "org": "Melady Lab @ USC",
7
- "testdata_leakage": "No"
8
- }
 
 
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/DC-research/TSorchestra",
6
  "org": "Melady Lab @ USC",
7
+ "testdata_leakage": "No",
8
+ "code_available": "No"
9
+ }
results/TTM-R1-Pretrained/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-ttm-r1",
6
  "org": "IBM Research",
7
- "testdata_leakage": "Yes"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-ttm-r1",
6
  "org": "IBM Research",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "Yes"
9
  }
results/TTM-R2-Finetuned/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/ttm.ipynb",
7
  "org": "IBM Research",
8
- "testdata_leakage": "Yes"
9
- }
 
 
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/ttm.ipynb",
7
  "org": "IBM Research",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
+ }
results/TTM-R2-Pretrained/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2",
6
  "org": "IBM Research",
7
- "testdata_leakage": "Yes"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-ttm-r2",
6
  "org": "IBM Research",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "Yes"
9
  }
results/TiRex/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/NX-AI/TiRex",
6
  "org": "NXAI",
7
- "testdata_leakage": "Yes"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/NX-AI/TiRex",
6
  "org": "NXAI",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "Yes"
9
  }
results/TimeCopilot/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "agentic",
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/AzulGarza/TimeCopilot",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "agentic",
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/AzulGarza/TimeCopilot",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/TimesFM-2.5/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/google/timesfm-2.5-200m-pytorch",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/timesfm2p5.ipynb",
7
  "org": "Google Research",
8
- "testdata_leakage": "No"
9
- }
 
 
5
  "model_link": "https://huggingface.co/google/timesfm-2.5-200m-pytorch",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/timesfm2p5.ipynb",
7
  "org": "Google Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
+ }
results/Toto_Open_Base_1.0/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/Datadog/Toto-Open-Base-1.0",
6
  "org": "Datadog",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/Datadog/Toto-Open-Base-1.0",
6
  "org": "Datadog",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/YingLong_110m/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_110m",
6
  "org": "Alibaba",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_110m",
6
  "org": "Alibaba",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/YingLong_300m/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_300m",
6
  "org": "Alibaba",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_300m",
6
  "org": "Alibaba",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/YingLong_50m/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_50m",
6
  "org": "Alibaba",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_50m",
6
  "org": "Alibaba",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/YingLong_6m/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_6m",
6
  "org": "Alibaba",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "bf16",
5
  "model_link": "https://huggingface.co/qcw2333/YingLong_6m",
6
  "org": "Alibaba",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/auto_arima/config.json CHANGED
@@ -2,5 +2,6 @@
2
  "model": "Auto_Arima",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
- "testdata_leakage": "No"
 
6
  }
 
2
  "model": "Auto_Arima",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
+ "testdata_leakage": "No",
6
+ "code_available": "Yes"
7
  }
results/auto_ets/config.json CHANGED
@@ -2,5 +2,6 @@
2
  "model": "Auto_ETS",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
- "testdata_leakage": "No"
 
6
  }
 
2
  "model": "Auto_ETS",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
+ "testdata_leakage": "No",
6
+ "code_available": "Yes"
7
  }
results/auto_theta/config.json CHANGED
@@ -2,5 +2,6 @@
2
  "model": "Auto_Theta",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
- "testdata_leakage": "No"
 
6
  }
 
2
  "model": "Auto_Theta",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
+ "testdata_leakage": "No",
6
+ "code_available": "Yes"
7
  }
results/chronos_base/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/amazon/chronos-t5-base",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
- "testdata_leakage": "Yes"
 
9
  }
 
5
  "model_link": "https://huggingface.co/amazon/chronos-t5-base",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
  }
results/chronos_bolt_base/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/amazon/chronos-bolt-base",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
- "testdata_leakage": "Yes"
 
9
  }
 
5
  "model_link": "https://huggingface.co/amazon/chronos-bolt-base",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
  }
results/chronos_bolt_small/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/amazon/chronos-bolt-small",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
- "testdata_leakage": "Yes"
 
9
  }
 
5
  "model_link": "https://huggingface.co/amazon/chronos-bolt-small",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
  }
results/chronos_large/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/amazon/chronos-t5-large",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
- "testdata_leakage": "Yes"
 
9
  }
 
5
  "model_link": "https://huggingface.co/amazon/chronos-t5-large",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/chronos.ipynb",
7
  "org": "AWS AI Labs",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
  }
results/crossformer/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Shanghai Jiao Tong University",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Shanghai Jiao Tong University",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/deepar/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Amazon Research",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Amazon Research",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/granite-flowstate-r1/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-flowstate-r1",
6
  "code_link": "https://github.com/ibm-granite/granite-tsfm/blob/main/notebooks/hfdemo/flowstate_gift_eval.ipynb",
7
  "org": "IBM Research",
8
- "testdata_leakage": "No"
 
9
  }
 
5
  "model_link": "https://huggingface.co/ibm-granite/granite-timeseries-flowstate-r1",
6
  "code_link": "https://github.com/ibm-granite/granite-tsfm/blob/main/notebooks/hfdemo/flowstate_gift_eval.ipynb",
7
  "org": "IBM Research",
8
+ "testdata_leakage": "No",
9
+ "code_available": "Yes"
10
  }
results/iTransformer/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Tsinghua University",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Tsinghua University",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/naive/config.json CHANGED
@@ -2,5 +2,6 @@
2
  "model": "Naive",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
- "testdata_leakage": "No"
 
6
  }
 
2
  "model": "Naive",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
+ "testdata_leakage": "No",
6
+ "code_available": "Yes"
7
  }
results/seasonal_naive/config.json CHANGED
@@ -2,5 +2,6 @@
2
  "model": "Seasonal_Naive",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
- "testdata_leakage": "No"
 
6
  }
 
2
  "model": "Seasonal_Naive",
3
  "model_type": "statistical",
4
  "model_dtype": "float32",
5
+ "testdata_leakage": "No",
6
+ "code_available": "Yes"
7
  }
results/sundial_base_128m/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/thuml/sundial-base-128m",
6
  "org": "Tsinghua University",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/thuml/sundial-base-128m",
6
  "org": "Tsinghua University",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/tabpfn_ts/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/liam-sbhoo/tabpfn-time-series/tree/main",
6
  "org": "PriorLabs",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/liam-sbhoo/tabpfn-time-series/tree/main",
6
  "org": "PriorLabs",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
results/tempo_ensemble/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/Melady/TEMPO",
6
  "org": "Melady Lab @ USC",
7
- "testdata_leakage": "Yes"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/Melady/TEMPO",
6
  "org": "Melady Lab @ USC",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "No"
9
  }
results/tft/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Google Research",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Google Research",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/tide/config.json CHANGED
@@ -3,5 +3,6 @@
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Google Research",
6
- "testdata_leakage": "No"
 
7
  }
 
3
  "model_type": "deep-learning",
4
  "model_dtype": "float32",
5
  "org": "Google Research",
6
+ "testdata_leakage": "No",
7
+ "code_available": "Yes"
8
  }
results/timesfm/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/google/timesfm-1.0-200m",
6
  "org": "Google Research",
7
- "testdata_leakage": "Yes"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://huggingface.co/google/timesfm-1.0-200m",
6
  "org": "Google Research",
7
+ "testdata_leakage": "Yes",
8
+ "code_available": "Yes"
9
  }
results/timesfm_2_0_500m/config.json CHANGED
@@ -5,5 +5,6 @@
5
  "model_link": "https://huggingface.co/google/timesfm-2.0-500m-jax",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/timesfm.ipynb",
7
  "org": "Google Research",
8
- "testdata_leakage": "Yes"
 
9
  }
 
5
  "model_link": "https://huggingface.co/google/timesfm-2.0-500m-jax",
6
  "code_link": "https://github.com/SalesforceAIResearch/gift-eval/blob/main/notebooks/timesfm.ipynb",
7
  "org": "Google Research",
8
+ "testdata_leakage": "Yes",
9
+ "code_available": "Yes"
10
  }
results/visionts/config.json CHANGED
@@ -4,5 +4,6 @@
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/Keytoyze/VisionTS",
6
  "org": "Zhejiang University",
7
- "testdata_leakage": "No"
 
8
  }
 
4
  "model_dtype": "float32",
5
  "model_link": "https://github.com/Keytoyze/VisionTS",
6
  "org": "Zhejiang University",
7
+ "testdata_leakage": "No",
8
+ "code_available": "Yes"
9
  }
src/display/utils.py CHANGED
@@ -35,7 +35,8 @@ model_info_dict.append(["params", ColumnContent, ColumnContent("#Params (B)", "n
35
  model_info_dict.append(["likes", ColumnContent, ColumnContent("Hub ❤️", "number", False, True)])
36
  model_info_dict.append(["still_on_hub", ColumnContent, ColumnContent("Available on the hub", "bool", False)])
37
  model_info_dict.append(["org", ColumnContent, ColumnContent("Organization", "str", True, hidden=False)])
38
- model_info_dict.append(["testdata_leakage", ColumnContent, ColumnContent("TestData Leakage", "str", True, hidden=False)])
 
39
  # model_info_dict.append(["revision", ColumnContent, ColumnContent("Model sha", "str", False, False)])
40
 
41
  # We use make dataclass to dynamically fill the scores from Tasks
 
35
  model_info_dict.append(["likes", ColumnContent, ColumnContent("Hub ❤️", "number", False, True)])
36
  model_info_dict.append(["still_on_hub", ColumnContent, ColumnContent("Available on the hub", "bool", False)])
37
  model_info_dict.append(["org", ColumnContent, ColumnContent("Organization", "str", True, hidden=False)])
38
+ model_info_dict.append(["testdata_leakage", ColumnContent, ColumnContent("Test Leak.", "str", True, hidden=False)])
39
+ model_info_dict.append(["code_available", ColumnContent, ColumnContent("OSS", "str", True, hidden=False)])
40
  # model_info_dict.append(["revision", ColumnContent, ColumnContent("Model sha", "str", False, False)])
41
 
42
  # We use make dataclass to dynamically fill the scores from Tasks
src/leaderboard/read_evals.py CHANGED
@@ -26,6 +26,7 @@ class ModelConfig:
26
  likes: int = 0
27
  num_params: int | str = 0
28
  testdata_leakage: str = "NA"
 
29
 
30
  @classmethod
31
  def init_from_json_file(cls, json_filepath):
@@ -43,8 +44,9 @@ class ModelConfig:
43
  code_link = data.get("code_link", "")
44
  org = data.get("org", "")
45
  testdata_leakage = data.get("testdata_leakage", "N/A")
 
46
  return cls(model=model, model_link=model_link, model_type=model_type, code_link=code_link, org=org,
47
- precision=precision, testdata_leakage=testdata_leakage)
48
 
49
  def to_dict(self):
50
  """Converts the model info to a dict compatible with our dataframe display"""
@@ -60,6 +62,7 @@ class ModelConfig:
60
  ModelInfoColumn.params.name: self.num_params,
61
  ModelInfoColumn.org.name: self.org,
62
  ModelInfoColumn.testdata_leakage.name: self.testdata_leakage,
 
63
  }
64
 
65
  return data_dict
 
26
  likes: int = 0
27
  num_params: int | str = 0
28
  testdata_leakage: str = "NA"
29
+ code_available: str = "NA"
30
 
31
  @classmethod
32
  def init_from_json_file(cls, json_filepath):
 
44
  code_link = data.get("code_link", "")
45
  org = data.get("org", "")
46
  testdata_leakage = data.get("testdata_leakage", "N/A")
47
+ code_available = data.get("code_available", "NA")
48
  return cls(model=model, model_link=model_link, model_type=model_type, code_link=code_link, org=org,
49
+ precision=precision, testdata_leakage=testdata_leakage, code_available=code_available)
50
 
51
  def to_dict(self):
52
  """Converts the model info to a dict compatible with our dataframe display"""
 
62
  ModelInfoColumn.params.name: self.num_params,
63
  ModelInfoColumn.org.name: self.org,
64
  ModelInfoColumn.testdata_leakage.name: self.testdata_leakage,
65
+ ModelInfoColumn.code_available.name: self.code_available,
66
  }
67
 
68
  return data_dict