Update README.md
Browse files
README.md
CHANGED
@@ -29,7 +29,9 @@ It combines classical deterministic error metrics with in-depth validations, inc
|
|
29 |
|
30 |
Moreover, tools for interpretability of DL models can be used to understand the inner workings and decision-making processes of these complex structures by analyzing the activations of different neurons and the importance of different features in the input data.
|
31 |
|
32 |
-
This work is funded by [Code for Earth 2023](https://codeforearth.ecmwf.int/) initiative.
|
|
|
|
|
33 |
|
34 |
|
35 |
|
@@ -41,10 +43,8 @@ This work is funded by [Code for Earth 2023](https://codeforearth.ecmwf.int/) in
|
|
41 |
- [Model Description](#model-description)
|
42 |
- [Uses](#uses)
|
43 |
- [Direct Use](#direct-use)
|
44 |
-
- [Downstream Use [Optional]](#downstream-use-optional)
|
45 |
- [Out-of-Scope Use](#out-of-scope-use)
|
46 |
- [Bias, Risks, and Limitations](#bias-risks-and-limitations)
|
47 |
-
- [Recommendations](#recommendations)
|
48 |
- [Training Details](#training-details)
|
49 |
- [Training Data](#training-data)
|
50 |
- [Training Procedure](#training-procedure)
|
@@ -56,8 +56,7 @@ This work is funded by [Code for Earth 2023](https://codeforearth.ecmwf.int/) in
|
|
56 |
- [Factors](#factors)
|
57 |
- [Metrics](#metrics)
|
58 |
- [Results](#results)
|
59 |
-
- [
|
60 |
-
- [Technical Specifications [optional]](#technical-specifications-optional)
|
61 |
- [Model Architecture and Objective](#model-architecture-and-objective)
|
62 |
- [Compute Infrastructure](#compute-infrastructure)
|
63 |
- [Hardware](#hardware)
|
@@ -69,7 +68,7 @@ This work is funded by [Code for Earth 2023](https://codeforearth.ecmwf.int/) in
|
|
69 |
## Model Description
|
70 |
|
71 |
<!-- Provide a longer summary of what this model is/does. -->
|
72 |
-
|
73 |
|
74 |
- **Developed by:** A team of Predictia Intelligent Data Solutions S.L. & Instituto de Fisica de Cantabria (IFCA)
|
75 |
- **Model type:** Vision model
|
@@ -83,6 +82,8 @@ A vision model for down-scaling (from 0.25º to 0.05º) regional reanalysis grid
|
|
83 |
|
84 |
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
85 |
|
|
|
|
|
86 |
|
87 |
|
88 |
## Out-of-Scope Use
|
@@ -100,9 +101,6 @@ A vision model for down-scaling (from 0.25º to 0.05º) regional reanalysis grid
|
|
100 |
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
|
101 |
|
102 |
|
103 |
-
## Recommendations
|
104 |
-
|
105 |
-
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
106 |
|
107 |
|
108 |
|
@@ -166,7 +164,8 @@ The data samples used for testing corresponds to the period from 2018 to 2020 (b
|
|
166 |
|
167 |
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
168 |
|
169 |
-
|
|
|
170 |
|
171 |
### Metrics
|
172 |
|
@@ -178,10 +177,6 @@ More information needed
|
|
178 |
|
179 |
More information needed
|
180 |
|
181 |
-
# Model Examination
|
182 |
-
|
183 |
-
More information needed
|
184 |
-
|
185 |
|
186 |
# Technical Specifications
|
187 |
|
|
|
29 |
|
30 |
Moreover, tools for interpretability of DL models can be used to understand the inner workings and decision-making processes of these complex structures by analyzing the activations of different neurons and the importance of different features in the input data.
|
31 |
|
32 |
+
This work is funded by [Code for Earth 2023](https://codeforearth.ecmwf.int/) initiative. The model **ConvSwin2SR** is released in Apache 2.0, making it usable without restrictions anywhere.
|
33 |
+
|
34 |
+
|
35 |
|
36 |
|
37 |
|
|
|
43 |
- [Model Description](#model-description)
|
44 |
- [Uses](#uses)
|
45 |
- [Direct Use](#direct-use)
|
|
|
46 |
- [Out-of-Scope Use](#out-of-scope-use)
|
47 |
- [Bias, Risks, and Limitations](#bias-risks-and-limitations)
|
|
|
48 |
- [Training Details](#training-details)
|
49 |
- [Training Data](#training-data)
|
50 |
- [Training Procedure](#training-procedure)
|
|
|
56 |
- [Factors](#factors)
|
57 |
- [Metrics](#metrics)
|
58 |
- [Results](#results)
|
59 |
+
- [Technical Specifications](#technical-specifications-optional)
|
|
|
60 |
- [Model Architecture and Objective](#model-architecture-and-objective)
|
61 |
- [Compute Infrastructure](#compute-infrastructure)
|
62 |
- [Hardware](#hardware)
|
|
|
68 |
## Model Description
|
69 |
|
70 |
<!-- Provide a longer summary of what this model is/does. -->
|
71 |
+
We present the ConvSwin2SR tranformer, a vision model for down-scaling (from 0.25º to 0.05º) regional reanalysis grids in the mediterranean area.
|
72 |
|
73 |
- **Developed by:** A team of Predictia Intelligent Data Solutions S.L. & Instituto de Fisica de Cantabria (IFCA)
|
74 |
- **Model type:** Vision model
|
|
|
82 |
|
83 |
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
84 |
|
85 |
+
## Direct Use
|
86 |
+
|
87 |
|
88 |
|
89 |
## Out-of-Scope Use
|
|
|
101 |
Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). Predictions generated by the model may include disturbing and harmful stereotypes across protected classes; identity characteristics; and sensitive, social, and occupational groups.
|
102 |
|
103 |
|
|
|
|
|
|
|
104 |
|
105 |
|
106 |
|
|
|
164 |
|
165 |
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
166 |
|
167 |
+
|
168 |
+
|
169 |
|
170 |
### Metrics
|
171 |
|
|
|
177 |
|
178 |
More information needed
|
179 |
|
|
|
|
|
|
|
|
|
180 |
|
181 |
# Technical Specifications
|
182 |
|