@ebgoldstein could you please help add / edit the info fields below for a Zoo model (I will revisit Fri)
Model Details.
Basic info
Person or organization developing model
Model date
Model version
Model type
Model files and their meaning / uses
json file containing model architecture
h5 file containing trained weights
model card (markdown format)
Information about training algorithms, parameters, fair-
ness constraints or other applied approaches, and features
Paper or other resource for more information
Citation details
License
Where to send questions or comments about the model - could be this Issues page
Intended Use.
Use cases that were envisioned during de-
velopment.
Primary intended uses: image segmentation with the intended image source and class list
Primary intended users: Zoo users / transfer learning
Out-of-scope use cases: whole image classification (does an image contain a class?) and object detection (masks can be converted into bounding boxes around each class)
Factors. Factors could include demographic or phenotypic
groups, environmental conditions, technical attributes, or
others listed in Section 4.3.
Relevant factors
Evaluation factors
IoU, Dice, KLD
Metrics. Metrics should be chosen to reflect potential real-
world impacts of the model.
Model performance measures
?
Decision thresholds
?
Variation approaches
?
Evaluation Data. Details on the dataset(s) used for the
quantitative analyses in the card.
Datasets
?
Motivation
?
Preprocessing
?
Training Data
description of training data
how many labeled images
how many classes / what classes
source of imagery
source of labels
link to dataset
e.g. C/T
• Quantitative Analyses
Unitary results
??
Intersectional results
??
Ethical Considerations
??
Caveats and Recommendations
A brief description of how to use this model with Segmentation Gym functions?
(this could/should be a Discussion, but that doesn't appear to be an option??)
I favour a simple text description of the models such as Fig 1 in https://arxiv.org/pdf/1810.03993.pdf in markdown format
@ebgoldstein could you please help add / edit the info fields below for a Zoo model (I will revisit Fri)
Model Details.
Intended Use.
Use cases that were envisioned during de- velopment.
Factors. Factors could include demographic or phenotypic
groups, environmental conditions, technical attributes, or others listed in Section 4.3.
Metrics. Metrics should be chosen to reflect potential real-
world impacts of the model.
Evaluation Data. Details on the dataset(s) used for the
quantitative analyses in the card.
Training Data
• Quantitative Analyses
Ethical Considerations
Caveats and Recommendations