File size: 13,110 Bytes
81ec35f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 |
<!DOCTYPE html>
<html>
<head>
<title>HTML Charset</title>
<meta charset="utf-8">
</head>
<body>
<h1>Model Name</h1>
<p> </p>
<p><strong>Section Overview:</strong> Provide the model name and a 1-2 sentence summary of what the model is.</p>
<p> </p>
<p><code>model_id</code></p>
<p> </p>
<p><code>model_summary</code></p>
<p> </p>
<h1> Table of Contents</h1>
<p> </p>
<p><strong>Section Overview:</strong> This section addresses questions around how the model is intended to be used in different applied contexts, discusses the foreseeable users of the model (including those affected by the model), and describes uses that are considered out of scope or misuse of the model. Note this section is not intended to include the license usage details. For that, link directly to the license.</p>
<p> </p>
<h1> Model Details</h1>
<p> </p>
<p><strong>Section Overview:</strong> This section provides basic information about what the model is, its current status, where it came from. It should be useful for anyone who wants to reference the model.</p>
<p> </p>
<h2><br></h2>
<h2> Model Description</h2>
<p> </p>
<p><code>model_description</code></p>
<p> </p>
<p><em>Provide details about the model. This includes the architecture, version, if it was introduced in a paper, if an implementation is available, and the creators. Any copyright should be attributed here. General information about training procedures, parameters, important disclaimers can also be mentioned in this section.</em></p>
<p> </p>
<ul>
<li><strong>Developed by:</strong> <code>developers</code></li>
</ul>
<p> </p>
<p><em>List (and ideally link to) the people who built the model.</em></p>
<p> </p>
<ul>
<li><strong>Funded by:</strong> <code>funded_by</code></li>
</ul>
<p> </p>
<p><em>List (and ideally link to) the funding sources that financially, computationally, or otherwise supported or enabled this model.</em></p>
<p> </p>
<ul>
<li><strong>Shared by [optional]:</strong> <code>shared_by</code></li>
</ul>
<p> </p>
<p><em>List (and ideally link to) the people/organization making the model available online.</em></p>
<p> </p>
<ul>
<li><strong>Model type:</strong> <code>model_type</code></li>
</ul>
<p> </p>
<p><em>You can name the “type” as:</em></p>
<p> </p>
<p><em>1. Supervision/Learning Method</em></p>
<p> </p>
<p><em>2. Machine Learning Type</em></p>
<p> </p>
<p><em>3. Modality</em></p>
<p> </p>
<ul>
<li><strong>Language(s)</strong> [NLP]: <code>language</code></li>
</ul>
<p> </p>
<p><em>Use this field when the system uses or processes natural (human) language..</em></p>
<p> </p>
<ul>
<li><strong>License:</strong> <code>license</code></li>
</ul>
<p> </p>
<p><em>Name and link to the license being used.</em></p>
<p> </p>
<ul>
<li><strong>Finetuned From Model [optional]:</strong> <code>base_model</code></li>
</ul>
<p> </p>
<p><em>** this model has another model as its base, link to that model here.</em></p>
<p> </p>
<h2><br></h2>
<h2> Model Sources optional</h2>
<p> </p>
<ul>
<li><strong>Repository:</strong> <code>repo</code></li>
<li> </li>
<li><strong>Paper [optional]:</strong> <code>paper</code></li>
<li> </li>
<li><strong>Demo [optional]:</strong> <code>demo</code></li>
</ul>
<p> </p>
<p><em>Provide sources for the user to see the model and its details. Additional kinds of resources – training logs, lessons learned, etc. – belong in the <a href="https://huggingface.co/docs/hub/en/model-card-annotated#more-information-optional">More Information</a> section. If you include one thing for this section, link to the repository.</em></p>
<p> </p>
<h1> Uses</h1>
<p> </p>
<p><strong>Section Overview:</strong> questions around how the model is intended to be used in different applied contexts, discusses the foreseeable users of the model (including those ... by the model). intended to include the license usage details. For that, link directly to the license.</p>
<p> </p>
<h2><br></h2>
<h2> Direct Use</h2>
<p> </p>
<p><code>direct_use</code></p>
<p> </p>
<p><em>Explain how the model can be used without fine-tuning, post-processing, or plugging into a pipeline. An example code snippet is recommended.</em></p>
<p> </p>
<h2><br></h2>
<h2> Downstream Use optional</h2>
<p> </p>
<p><code>downstream_use</code></p>
<p> </p>
<p><em>Explain how this model can be used and fine-tuned for a task or when plugged into a larger ecosystem or app. An example code snippet is recommended.</em></p>
<p> </p>
<h2><br></h2>
<h2> Out-of-Scope Use</h2>
<p> </p>
<p><code>out_of_scope_use</code></p>
<p> </p>
<p><em>List how the model may foreseeably be misused (used in a way it will not work for) and address what users ought not do with the model.</em></p>
<p> </p>
<h1> Bias, Risks, and Limitations</h1>
<p> </p>
<p><strong>Section Overview:</strong> This section identifies harms, misunderstandings, and technical and sociotechnical limitations. It also provides potential mitigations. Bias, risks, and limitations can sometimes be inseparable/refer to the same issues. Generally, bias and risks are sociotechnical, while limitations are technical:</p>
<p> </p>
<ul>
<li>A <strong>bias</strong> is a stereotype or disproportionate performance (skew) for some subpopulations.</li>
<li> </li>
<li>A <strong>risk</strong> is a socially-sensitive issue that the model might cause.</li>
<li> </li>
<li>A <strong>limitation</strong> is a likely failure to be addressed following the listed Recommendations.</li>
</ul>
<p> </p>
<p><code>bias_risks_limitations</code></p>
<p> </p>
<p><em>What are the known or foreseeable issues stemming from this model?</em></p>
<p> </p>
<h2><br></h2>
<h2> Recommendations</h2>
<p> </p>
<p><code>bias_recommendations</code></p>
<p> </p>
<p><em>What are recommendations with respect to the foreseeable issues? This can include everything from “downsample your image” to filtering explicit content..</em></p>
<p> </p>
<h1> Training Details</h1>
<p> </p>
<p><strong>Section Overview:</strong> This section provides information to describe and replicate training, including the training data, the speed and size of training elements, and the environmental impact of training. <a href="https://huggingface.co/docs/hub/en/model-card-annotated#technical-specifications-optional">Technical Specifications</a> as well, and content here should link to that section when it is relevant to the training procedure. useful for people who want to learn more about the model inputs training footprint. for anyone who wants to know the basics of what the model is learning.</p>
<p> </p>
<h2><br></h2>
<h2> Training Data</h2>
<p> </p>
<p><code>training_data</code></p>
<p> </p>
<p><em>Write 1-2 sentences related to data pre-processing or additional filtering <a href="https://huggingface.co/docs/hub/en/model-card-annotated#more-information-optional">More Information</a>.</em></p>
<p> </p>
<h2><br></h2>
<h2> Procedure optional</h2>
<p> </p>
<h3><br></h3>
<h3> Preprocessing</h3>
<p> </p>
<p><code>preprocessing</code></p>
<p> </p>
<p><em>Detail tokenization, resizing/rewriting (depending on the modality), etc.</em></p>
<p> </p>
<h3><br></h3>
<h3> Speeds, Sizes, Times</h3>
<p> </p>
<p><code>speeds_sizes_times</code></p>
<p> </p>
<p><em>Detail throughput, start/end time, checkpoint sizes, etc.</em></p>
<p> </p>
<h1> Evaluation</h1>
<p> </p>
<p><strong>Section Overview:</strong> evaluation protocols. Target fairness metrics should be decided based on errors are more likely to be identified in light of the model use. specify model’s evaluation results in a structured way in the model card metadata. parsed and displayed in a widget on the model page. See <a href="https://huggingface.co/docs/hub/model-cards#evaluation-results" rel="nofollow">https://huggingface.co/docs/hub/model-cards#evaluation-results</a>.</p>
<p> </p>
<h2><br></h2>
<h2> Data, Factors & Metrics</h2>
<p> </p>
<h3><br></h3>
<h3> Testing Data</h3>
<p> </p>
<p><code>testing_data</code></p>
<p> </p>
<p><em>Ideally this links to a Dataset Card for testing data.</em></p>
<p> </p>
<h3><br></h3>
<h3> Factors</h3>
<p> </p>
<p><code>testing_factors</code></p>
<p> </p>
<p><em>What are the foreseeable circumstances that will influence how the model behaves? This includes domain and context, as well as population subgroups. Evaluation should ideally be <strong>disaggregated</strong> across factors in order to uncover disparities in performance.</em></p>
<p> </p>
<h3><br></h3>
<h3> Metrics</h3>
<p> </p>
<p><code>testing_metrics</code></p>
<p> </p>
<p><em> metrics for evaluation in light of tradeoffs between different errors?</em></p>
<p> </p>
<h2><br></h2>
<h2> Results</h2>
<p> </p>
<p><code>results</code></p>
<p> </p>
<p><em>Results based on the Factors and Metrics defined above.</em></p>
<p> </p>
<h3><br></h3>
<h3> Summary</h3>
<p> </p>
<p><code>results_summary</code></p>
<p> </p>
<p><em>What do the results say? This can function as a kind of tl;dr for general audiences..</em></p>
<p> </p>
<h1> Model Examination optional</h1>
<p> </p>
<p><strong>Section Overview:</strong> examination</p>
<p> </p>
<p><code>model_examination</code></p>
<p> </p>
<h1> Environmental Impact</h1>
<p> </p>
<p><strong>Section Overview:</strong> Summarizes the information necessary to calculate environmental impacts .</p>
<p> </p>
<ul>
<li><strong>Hardware Type:</strong> <code>hardware_type</code></li>
<li> </li>
<li><strong>Hours used:</strong> <code>hours_used</code></li>
<li> </li>
<li><strong>Cloud Provider:</strong> <code>cloud_provider</code></li>
<li> </li>
<li><strong>Compute Region:</strong> <code>cloud_region</code></li>
<li> </li>
<li><strong>Carbon Emitted:</strong> <code>co2_emitted</code></li>
</ul>
<p> </p>
<p><em>Carbon emissions can be estimated using the <a href="https://mlco2.github.io/impact#compute" rel="nofollow">Machine Learning Impact calculator</a> presented in <a href="https://arxiv.org/abs/1910.09700" rel="nofollow">source</a>.</em></p>
<p> </p>
<h1> Technical Specifications optional</h1>
<p> </p>
<p><strong>Section Overview:</strong> This section includes details about the model architecture, and the compute infrastructure. </p>
<p> </p>
<h2><br></h2>
<h2> Model Architecture and Objective</h2>
<p> </p>
<p><code>model_specs</code></p>
<p> </p>
<h2><br></h2>
<h2> Compute Infrastructure</h2>
<p> </p>
<p><code>compute_infrastructure</code></p>
<p> </p>
<h3><br></h3>
<h3> Hardware</h3>
<p> </p>
<p><code>hardware_requirements</code></p>
<p> </p>
<p><em>What are the minimum hardware requirements, e.g. processing, storage, and memory requirements?</em></p>
<p> </p>
<h3><br></h3>
<h3> Software</h3>
<p> </p>
<p><code>software</code></p>
<p> </p>
<h1> optional</h1>
<p> </p>
<p><strong>Section Overview:</strong> The developers’ preferred citation for this model. </p>
<p> </p>
<h3><br></h3>
<h3> BibTeX</h3>
<p> </p>
<p><code>citation_bibtex</code></p>
<p> </p>
<h3><br></h3>
<h3> APA</h3>
<p> </p>
<p><code>citation_apa</code></p>
<p> </p>
<h1> Glossary optional</h1>
<p> </p>
<p><strong>Section Overview:</strong> This section defines common terms and how metrics are calculated.</p>
<p> </p>
<p><code>glossary</code></p>
<p> </p>
<p><em>Clearly define terms in order to be accessible across audiences.</em></p>
<p> </p>
<h1> More Information optional</h1>
<p> </p>
<p><strong>Section Overview:</strong> lessons learned and more .</p>
<p> </p>
<p><code>more_information</code></p>
<p> </p>
<h1> Model Card Authors optional</h1>
<p> </p>
<p><strong>Section Overview:</strong> who create the model card, .</p>
<p> </p>
<p><code>model_card_authors</code></p>
<p> </p>
<h1> Model Card Contact</h1>
<p> </p>
<p><strong>Section Overview:</strong> contact</p>
<p> </p>
<p><code>model_card_contact</code></p>
<p> </p>
<div>
<h1> How to Get Started with the Model</h1> <p><strong>Section Overview:</strong> Provides a code snippet to show how to use the model.</p> <p><code>get_started_code</code></p> <p><br></p> <div aria-live="assertive" aria-atomic="true"><br></div>
</div>
<p> </p>
<div>
<div><a href="https://huggingface.co/docs/hub/en/model-cards"><span>←</span>Model Cards</a></div>
</div>
</body>
</html> |