Update README.md
Browse files
README.md
CHANGED
@@ -47,59 +47,59 @@ trunk
|
|
47 |
|
48 |
|
49 |
- **Developed by:** M. Maruf and Anuj Karpatne
|
50 |
-
|
51 |
-
|
52 |
-
|
53 |
-
- **License:**
|
54 |
-
|
55 |
|
56 |
-
### Model Sources
|
57 |
|
58 |
<!-- Provide the basic links for the model. -->
|
59 |
|
60 |
- **Repository:** [BGNN-trait-segmentation](https://github.com/hdr-bgnn/BGNN-trait-segmentation)
|
61 |
-
- **Paper
|
62 |
-
|
63 |
|
64 |
-
## Uses
|
65 |
|
66 |
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
67 |
|
68 |
-
### Direct Use
|
69 |
|
70 |
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
71 |
|
72 |
-
[More Information Needed]
|
73 |
|
74 |
-
### Downstream Use [optional]
|
75 |
|
76 |
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
77 |
|
78 |
-
[More Information Needed]
|
79 |
|
80 |
-
### Out-of-Scope Use
|
81 |
|
82 |
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
83 |
|
84 |
-
[More Information Needed]
|
85 |
|
86 |
-
## Bias, Risks, and Limitations
|
87 |
|
88 |
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
89 |
|
90 |
-
[More Information Needed]
|
91 |
|
92 |
-
### Recommendations
|
93 |
|
94 |
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
95 |
|
96 |
-
Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
|
97 |
|
98 |
## How to Get Started with the Model
|
99 |
|
100 |
<!--Use the code below to get started with the model.-->
|
101 |
|
102 |
-
See instructions for use [here](https://github.com/hdr-bgnn/BGNN-trait-segmentation/blob/main/Segment_mini/README.md)
|
103 |
|
104 |
## Training Details
|
105 |
|
@@ -138,20 +138,20 @@ We then trained the prepared model for 120 epochs, updating the weights using di
|
|
138 |
The Adam optimizer ([Kingma & Ba, 2014](https://doi.org/10.48550/arXiv.1412.6980)) with a small learning rate (1e-4) was used to update the model weights.
|
139 |
|
140 |
|
141 |
-
#### Preprocessing [optional]
|
142 |
|
143 |
-
[More Information Needed]
|
144 |
|
145 |
|
146 |
-
#### Training Hyperparameters
|
147 |
|
148 |
-
- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
149 |
|
150 |
-
#### Speeds, Sizes, Times [optional]
|
151 |
|
152 |
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
153 |
|
154 |
-
[More Information Needed]
|
155 |
|
156 |
## Evaluation
|
157 |
|
@@ -164,87 +164,87 @@ Our segmentation model achieved a 0.90 mIoU score on the test dataset.
|
|
164 |
|
165 |
We had 99 testing images and 98 validation images.
|
166 |
|
167 |
-
#### Testing Data
|
168 |
|
169 |
<!-- This should link to a Data Card if possible. -->
|
170 |
|
171 |
-
[More Information Needed]
|
172 |
|
173 |
-
#### Factors
|
174 |
|
175 |
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
176 |
|
177 |
-
[More Information Needed]
|
178 |
|
179 |
-
#### Metrics
|
180 |
|
181 |
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
182 |
|
183 |
-
[More Information Needed]
|
184 |
|
185 |
-
### Results
|
186 |
|
187 |
-
[More Information Needed]
|
188 |
|
189 |
|
190 |
|
191 |
-
#### Summary
|
192 |
|
193 |
|
194 |
|
195 |
-
## Model Examination [optional]
|
196 |
|
197 |
<!-- Relevant interpretability work for the model goes here -->
|
198 |
|
199 |
-
[More Information Needed]
|
200 |
|
201 |
-
## Environmental Impact
|
202 |
|
203 |
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
204 |
|
205 |
-
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
206 |
|
207 |
- **Hardware Type:** [More Information Needed]
|
208 |
- **Hours used:** [More Information Needed]
|
209 |
- **Cloud Provider:** [More Information Needed]
|
210 |
- **Compute Region:** [More Information Needed]
|
211 |
-
- **Carbon Emitted:** [More Information Needed]
|
212 |
|
213 |
-
## Technical Specifications [optional]
|
214 |
|
215 |
-
### Model Architecture and Objective
|
216 |
|
217 |
-
[More Information Needed]
|
218 |
|
219 |
-
### Compute Infrastructure
|
220 |
|
221 |
-
[More Information Needed]
|
222 |
|
223 |
-
#### Hardware
|
224 |
|
225 |
-
[More Information Needed]
|
226 |
|
227 |
-
#### Software
|
228 |
|
229 |
-
[More Information Needed]
|
230 |
|
231 |
-
## Citation [optional]
|
232 |
|
233 |
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
234 |
-
|
235 |
-
**BibTeX:**
|
236 |
|
237 |
[More Information Needed]
|
238 |
|
239 |
**APA:**
|
240 |
|
241 |
-
[More Information Needed]
|
242 |
|
243 |
-
## Glossary [optional]
|
244 |
|
245 |
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
246 |
|
247 |
-
[More Information Needed]
|
248 |
|
249 |
## More Information [optional]
|
250 |
|
@@ -256,6 +256,6 @@ Carbon emissions can be estimated using the [Machine Learning Impact calculator]
|
|
256 |
|
257 |
## Model Card Contact
|
258 |
|
259 |
-
[More Information Needed]
|
260 |
|
261 |
|
|
|
47 |
|
48 |
|
49 |
- **Developed by:** M. Maruf and Anuj Karpatne
|
50 |
+
<!--- **Shared by [optional]:** [More Information Needed]-->
|
51 |
+
<!--- **Model type:** [More Information Needed]-->
|
52 |
+
<!--- **Language(s) (NLP):** [More Information Needed]-->
|
53 |
+
- **License:** MIT
|
54 |
+
<!--- **Finetuned from model [optional]:** [More Information Needed]-->
|
55 |
|
56 |
+
### Model Sources
|
57 |
|
58 |
<!-- Provide the basic links for the model. -->
|
59 |
|
60 |
- **Repository:** [BGNN-trait-segmentation](https://github.com/hdr-bgnn/BGNN-trait-segmentation)
|
61 |
+
- **Paper:** [In progress]
|
62 |
+
<!--- **Demo [optional]:** [More Information Needed]-->
|
63 |
|
64 |
+
<!-- ## Uses
|
65 |
|
66 |
<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
|
67 |
|
68 |
+
<!-- ### Direct Use
|
69 |
|
70 |
<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
|
71 |
|
72 |
+
<!-- [More Information Needed]
|
73 |
|
74 |
+
<!-- ### Downstream Use [optional]
|
75 |
|
76 |
<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
|
77 |
|
78 |
+
<!-- [More Information Needed]
|
79 |
|
80 |
+
<!-- ### Out-of-Scope Use
|
81 |
|
82 |
<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
|
83 |
|
84 |
+
<!-- [More Information Needed]
|
85 |
|
86 |
+
<!-- ## Bias, Risks, and Limitations
|
87 |
|
88 |
<!-- This section is meant to convey both technical and sociotechnical limitations. -->
|
89 |
|
90 |
+
<!-- [More Information Needed] -->
|
91 |
|
92 |
+
<!-- ### Recommendations -->
|
93 |
|
94 |
<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
|
95 |
|
96 |
+
<!-- Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations. -->
|
97 |
|
98 |
## How to Get Started with the Model
|
99 |
|
100 |
<!--Use the code below to get started with the model.-->
|
101 |
|
102 |
+
See instructions for use [here](https://github.com/hdr-bgnn/BGNN-trait-segmentation/blob/main/Segment_mini/README.md).
|
103 |
|
104 |
## Training Details
|
105 |
|
|
|
138 |
The Adam optimizer ([Kingma & Ba, 2014](https://doi.org/10.48550/arXiv.1412.6980)) with a small learning rate (1e-4) was used to update the model weights.
|
139 |
|
140 |
|
141 |
+
<!-- #### Preprocessing [optional]
|
142 |
|
143 |
+
<!-- [More Information Needed]
|
144 |
|
145 |
|
146 |
+
<!-- #### Training Hyperparameters
|
147 |
|
148 |
+
<!-- - **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
|
149 |
|
150 |
+
<!-- #### Speeds, Sizes, Times [optional]
|
151 |
|
152 |
<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
|
153 |
|
154 |
+
<!-- [More Information Needed] -->
|
155 |
|
156 |
## Evaluation
|
157 |
|
|
|
164 |
|
165 |
We had 99 testing images and 98 validation images.
|
166 |
|
167 |
+
<!-- #### Testing Data
|
168 |
|
169 |
<!-- This should link to a Data Card if possible. -->
|
170 |
|
171 |
+
<!-- [More Information Needed]
|
172 |
|
173 |
+
<!-- #### Factors
|
174 |
|
175 |
<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
|
176 |
|
177 |
+
<!-- [More Information Needed]
|
178 |
|
179 |
+
<!-- #### Metrics
|
180 |
|
181 |
<!-- These are the evaluation metrics being used, ideally with a description of why. -->
|
182 |
|
183 |
+
<!-- [More Information Needed]
|
184 |
|
185 |
+
<!-- ### Results
|
186 |
|
187 |
+
<!-- [More Information Needed]
|
188 |
|
189 |
|
190 |
|
191 |
+
<!-- #### Summary
|
192 |
|
193 |
|
194 |
|
195 |
+
<!-- ## Model Examination [optional]
|
196 |
|
197 |
<!-- Relevant interpretability work for the model goes here -->
|
198 |
|
199 |
+
<!-- [More Information Needed]
|
200 |
|
201 |
+
<!-- ## Environmental Impact
|
202 |
|
203 |
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
|
204 |
|
205 |
+
<!-- Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
|
206 |
|
207 |
- **Hardware Type:** [More Information Needed]
|
208 |
- **Hours used:** [More Information Needed]
|
209 |
- **Cloud Provider:** [More Information Needed]
|
210 |
- **Compute Region:** [More Information Needed]
|
211 |
+
- **Carbon Emitted:** [More Information Needed] -->
|
212 |
|
213 |
+
<!-- ## Technical Specifications [optional]
|
214 |
|
215 |
+
<!-- ### Model Architecture and Objective
|
216 |
|
217 |
+
<!-- [More Information Needed]
|
218 |
|
219 |
+
<!-- ### Compute Infrastructure
|
220 |
|
221 |
+
<!-- [More Information Needed]
|
222 |
|
223 |
+
<!-- #### Hardware
|
224 |
|
225 |
+
<!-- [More Information Needed]
|
226 |
|
227 |
+
<!-- #### Software
|
228 |
|
229 |
+
<!-- [More Information Needed] -->
|
230 |
|
231 |
+
<!-- ## Citation [optional]
|
232 |
|
233 |
<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
|
234 |
+
<!-- M. Maruf, 2023, "BGNN-trait-segmentation", DOI coming with model soon for paper citation -->
|
235 |
+
<!-- **BibTeX:**
|
236 |
|
237 |
[More Information Needed]
|
238 |
|
239 |
**APA:**
|
240 |
|
241 |
+
[More Information Needed] -->
|
242 |
|
243 |
+
<!-- ## Glossary [optional]
|
244 |
|
245 |
<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
|
246 |
|
247 |
+
<!-- [More Information Needed]
|
248 |
|
249 |
## More Information [optional]
|
250 |
|
|
|
256 |
|
257 |
## Model Card Contact
|
258 |
|
259 |
+
[More Information Needed] -->
|
260 |
|
261 |
|