English
kumo-ts
adgautam commited on
Commit
b15386b
·
verified ·
1 Parent(s): abff20a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -2,12 +2,12 @@
2
  license: apache-2.0
3
  language:
4
  - en
5
- library_name: nv-tesseract
6
  ---
7
- # NV-Tesseract-Forecasting Overview
8
 
9
  ## Description: <br>
10
- NVIDIA NV-Tesseract-Forecasting provides forecasting functionality via NVIDIA's novel DARR (Domain-Aware Representation and Retrieval) which is a context-enhanced forecasting mode that combines direct model predictions with pattern retrieval from historical data. The model leverages MOMENT encoders with a trainable forecasting head for accurate long-horizon time series predictions.
11
 
12
  This model is for research and development only.
13
 
@@ -22,7 +22,7 @@ Global <br>
22
  Companies, organizations, research hubs looking to do long/short-horizon forecasting on temporal data. <br>
23
 
24
  ### Release Date: <br>
25
- HuggingFace 04/07/2026 via https://huggingface.co/nvidia/nv-tesseract-forecasting
26
 
27
  ## Reference(s):
28
  <a href="https://ieeexplore.ieee.org/abstract/document/11402442">Memory-Augmented Forecasting: Scalability and Generalization Across Temporal Domains</a> </br>
@@ -37,7 +37,7 @@ HuggingFace 04/07/2026 via https://huggingface.co/nvidia/nv-tesseract-forecastin
37
 
38
  **Number of model parameters:** ~340 million (base encoder) + trainable forecasting head <br>
39
 
40
- The NV-Tesseract-Forecasting is a Transformer-based model for time series forecasting. It uses MOMENT encoders with frozen encoder and embedder weights, while training only the forecasting head. The model supports DARR (Domain-Aware Representation and Retrieval) mode, which enhances predictions by combining direct forecasts with kNN-based pattern retrieval from historical context data. <br>
41
 
42
  ## Input: <br>
43
  **Input Type(s):** Tabular numeric <br>
@@ -69,7 +69,7 @@ Linux<br>
69
  The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.
70
 
71
  ## Model Version(s):
72
- NV-Tesseract-Forecasting<br>
73
 
74
  # Training & Testing Datasets: <br>
75
  80/20 split per dataset. <br />
 
2
  license: apache-2.0
3
  language:
4
  - en
5
+ library_name: kumo-ts
6
  ---
7
+ # Kumo-Forecast Overview
8
 
9
  ## Description: <br>
10
+ NVIDIA Kumo-Forecast provides forecasting functionality via NVIDIA's novel DARR (Domain-Aware Representation and Retrieval) which is a context-enhanced forecasting mode that combines direct model predictions with pattern retrieval from historical data. The model leverages MOMENT encoders with a trainable forecasting head for accurate long-horizon time series predictions.
11
 
12
  This model is for research and development only.
13
 
 
22
  Companies, organizations, research hubs looking to do long/short-horizon forecasting on temporal data. <br>
23
 
24
  ### Release Date: <br>
25
+ HuggingFace 04/07/2026 via https://huggingface.co/nvidia/Kumo-Forecast
26
 
27
  ## Reference(s):
28
  <a href="https://ieeexplore.ieee.org/abstract/document/11402442">Memory-Augmented Forecasting: Scalability and Generalization Across Temporal Domains</a> </br>
 
37
 
38
  **Number of model parameters:** ~340 million (base encoder) + trainable forecasting head <br>
39
 
40
+ The Kumo-Forecast is a Transformer-based model for time series forecasting. It uses MOMENT encoders with frozen encoder and embedder weights, while training only the forecasting head. The model supports DARR (Domain-Aware Representation and Retrieval) mode, which enhances predictions by combining direct forecasts with kNN-based pattern retrieval from historical context data. <br>
41
 
42
  ## Input: <br>
43
  **Input Type(s):** Tabular numeric <br>
 
69
  The integration of foundation and fine-tuned models into AI systems requires additional testing using use-case-specific data to ensure safe and effective deployment. Following the V-model methodology, iterative testing and validation at both unit and system levels are essential to mitigate risks, meet technical and functional requirements, and ensure compliance with safety and ethical standards before deployment.
70
 
71
  ## Model Version(s):
72
+ Kumo-Forecast<br>
73
 
74
  # Training & Testing Datasets: <br>
75
  80/20 split per dataset. <br />