Update README.md
Browse files
README.md
CHANGED
|
@@ -135,3 +135,30 @@ forecast_df = tfm.forecast_on_df(
|
|
| 135 |
num_jobs=-1,
|
| 136 |
)
|
| 137 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 135 |
num_jobs=-1,
|
| 136 |
)
|
| 137 |
```
|
| 138 |
+
|
| 139 |
+
## Pretraining Data
|
| 140 |
+
|
| 141 |
+
It is important to list all the data sources in order to enable fair benchmarking. The TimesFM 2.0 series contains the pretraining set of TimesFM 1.0 along with these additional datasets:
|
| 142 |
+
|
| 143 |
+
| Dataset | Download Source |
|
| 144 |
+
| :------------------------------------------ | :----------------------- |
|
| 145 |
+
| azure_vm_traces | LOTSA Pretrain |
|
| 146 |
+
| residential_load_power | LOTSA Pretrain |
|
| 147 |
+
| borg_cluster_data | LOTSA Pretrain |
|
| 148 |
+
| residential_pv_power | LOTSA Pretrain |
|
| 149 |
+
| q_traffic | LOTSA Pretrain |
|
| 150 |
+
| london_smart_meters_with_missing | LOTSA Pretrain |
|
| 151 |
+
| taxi_30min | LOTSA Pretrain |
|
| 152 |
+
| solar_power | LOTSA Pretrain |
|
| 153 |
+
| wind_power | LOTSA Pretrain |
|
| 154 |
+
| kdd2022 | LOTSA Pretrain |
|
| 155 |
+
| largest | LOTSA Pretrain |
|
| 156 |
+
| era5 | LOTSA Pretrain |
|
| 157 |
+
| buildings | LOTSA Pretrain |
|
| 158 |
+
| cmip6 | LOTSA Pretrain |
|
| 159 |
+
| china_air_quality | LOTSA Pretrain |
|
| 160 |
+
| beijing_air_quality | LOTSA Pretrain |
|
| 161 |
+
| subseasonal | LOTSA Pretrain |
|
| 162 |
+
| kaggle_web_traffic_weekly | LOTSA Pretrain |
|
| 163 |
+
| cdc_fluview_who_nrevss | LOTSA Pretrain |
|
| 164 |
+
| godaddy | LOTSA Pretrain |
|