Update README.md

README.md

@@ -55,56 +55,25 @@ getting started [notebook](https://github.com/IBM/tsfm/blob/main/notebooks/hfdem
 ## Model Releases (along with the branch name where the models are stored):
 
 
-- **512-96-r2**: Given the last 512 time-points (i.e. context length), this model can forecast up to next 96 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: main) 
-- **1024-96-r2**: Given the last 1024 time-points (i.e. context length), this model can forecast up to next 96 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1024-96-r2) 
-
-- **1536-96-r2**: Given the last 1536 time-points (i.e. context length), this model can forecast up to next 96 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1536-96-r2) 
-
-- **512-192-r2**: Given the last 512 time-points (i.e. context length), this model can forecast up to next 192 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 512-192-r2) 
-  
-- **1024-192-r2**: Given the last 1024 time-points (i.e. context length), this model can forecast up to next 192 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1024-192-r2) 
 
-- **1536-192-r2**: Given the last 1536 time-points (i.e. context length), this model can forecast up to next 192 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1536-192-r2)
 
+- **512-96-r2**: Given the last 512 time-points (i.e. context length), this model can forecast up to the next 96 time-points (i.e. forecast length)
+    in future. (branch name: main)
   
-- **512-336-r2**: Given the last 512 time-points (i.e. context length), this model can forecast up to next 336 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 512-336-r2) 
-  
-- **1024-336-r2**: Given the last 1024 time-points (i.e. context length), this model can forecast up to next 336 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1024-336-r2) 
+- **1024-96-r2**: Given the last 1024 time-points (i.e. context length), this model can forecast up to the next 96 time-points (i.e. forecast length)
+    in future.  (branch name: 1024-96-r2) [[Benchmarks]]
 
-- **1536-336-r2**: Given the last 1536 time-points (i.e. context length), this model can forecast up to next 336 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1536-336-r2)
+- **1536-96-r2**: Given the last 1536 time-points (i.e. context length), this model can forecast up to the next 96 time-points (i.e. forecast length)
+    in future. (branch name: 1536-96-r2)
 
+- Likewise, we have models released for forecast lengths up to 720 timepoints. The branch names for these are as follows: `512-192-r2`, `1024-192-r2`, `1536-192-r2`, `512-336-r2`,
+  `512-336-r2`, `1024-336-r2`, `1536-336-r2`, `512-720-r2`, `1024-720-r2`, `1536-720-r2`
 
-  
-- **512-720-r2**: Given the last 512 time-points (i.e. context length), this model can forecast up to next 720 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 512-720-r2) 
-  
-- **1024-720-r2**: Given the last 1024 time-points (i.e. context length), this model can forecast up to next 720 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1024-720-r2) 
+- Please use the [[get_model]](https://github.com/ibm-granite/granite-tsfm/blob/main/tsfm_public/toolkit/get_model.py) utility to automatically select the required model based on your input context length and forecast length requirement.
+
+- We currently allow 3 context lengths (512, 1024 and 1536) and 4 forecast lengths (96, 192, 336, 720). Users need to provide one of the 3 allowed context lengths as input.
+but can provide any forecast lengths up to 720 in get_model() to get the required model.
 
-- **1536-720-r2**: Given the last 1536 time-points (i.e. context length), this model can forecast up to next 720 time-points (i.e. forecast length)
-    in future. This model is pre-trained with a larger pretraining dataset for improved accuracy. Recommended for hourly and minutely
-    resolutions (Ex. 10 min, 15 min, 1 hour, etc).   (branch name: 1536-720-r2) 
-    
 
 ## Model Capabilities with example scripts
 
@@ -118,11 +87,15 @@ The below model scripts can be used for any of the above TTM models. Please upda
 - **New Releases (extended features released on October 2024)**
   - Finetuning and Forecasting with Exogenous/Control Variables [[Example]](https://github.com/ibm-granite/granite-tsfm/blob/main/notebooks/tutorial/ttm_with_exog_tutorial.ipynb)
   - Finetuning and Forecasting with static categorical features [Example: To be added soon]
-  - Rolling Forecasts - Extend forecast lengths beyond 96 via rolling capability [[Example]](https://github.com/ibm-granite/granite-tsfm/blob/main/notebooks/hfdemo/ttm_rolling_prediction_getting_started.ipynb)
+  - Rolling Forecasts - Extend forecast lengths via rolling capability. Rolling beyond 2*forecast_length is not recommended. [[Example]](https://github.com/ibm-granite/granite-tsfm/blob/main/notebooks/hfdemo/ttm_rolling_prediction_getting_started.ipynb)
   - Helper scripts for optimal Learning Rate suggestions for Finetuning [[Example]](https://github.com/ibm-granite/granite-tsfm/blob/main/notebooks/tutorial/ttm_with_exog_tutorial.ipynb)
     
 ## Benchmarks
 
+<p align="center" width="100%">
+<img src="benchmarks.webp" width="600">
+</p>
+
 TTM outperforms popular benchmarks such as TimesFM, Moirai, Chronos, Lag-Llama, Moment, GPT4TS, TimeLLM, LLMTime in zero/fewshot forecasting while reducing computational requirements significantly. 
 Moreover, TTMs are lightweight and can be executed even on CPU-only machines, enhancing usability and fostering wider
 adoption in resource-constrained environments. For more details, refer to our [paper](https://arxiv.org/pdf/2401.03955.pdf).
@@ -130,6 +103,12 @@ adoption in resource-constrained environments. For more details, refer to our [p
 - TTM-E referred in the paper maps to the 1024 context models.
 - TTM-A referred in the paper maps to the 1536 context models.
 
+Please note that the Granite TTM models are pre-trained exclusively on datasets
+with clear commercial-use licenses that are approved by our legal team. As a result, the pre-training dataset used in this release differs slightly from the one used in the research
+paper, which may lead to minor variations in model performance as compared to the published results. Please refer to our paper for more details.
+
+**Benchmarking Scripts: [here](https://github.com/ibm-granite/granite-tsfm/tree/main/notebooks/hfdemo/tinytimemixer/full_benchmarking)**
+
 
 ## Recommended Use
 1. Users have to externally standard scale their data independently for every channel before feeding it to the model (Refer to [TSP](https://github.com/IBM/tsfm/blob/main/tsfm_public/toolkit/time_series_preprocessor.py), our data processing utility for data scaling.)