Use first level header for "Model Card", remove extraneous whitespace (#2)

Browse files

- Use first level header for "Model Card", remove extraneous whitespace (bf703f814681b289c4663fa0fff699b605c402e0)

Files changed (1) hide show

README.md +2 -10

README.md CHANGED Viewed

@@ -6,7 +6,7 @@ base_model:
 library_name: sentence-transformers
 ---
 Model Card
-----------
 _Who to contact:_ fbda [at] nfi [dot] nl \
 TODO: add link to github repo
@@ -22,7 +22,6 @@ The model architecture is inspired by [jTrans](https://github.com/vul337/jTrans)
 although the typical Next Sentence Prediction has been replaced with Jump Target Prediction, as proposed in Wang et al.
 This architecture has subsequently been finetuned for semantic search purposes. We have followed the procedure proposed by [S-BERT](https://www.sbert.net/examples/applications/semantic-search/README.html).
 ### What is the output of the model?
 The model returns a vector of 768 dimensions for each function that it's given. These vectors can be compared to
 get an indication of which functions are similar to each other.
@@ -33,13 +32,11 @@ The model has been evaluated on [Mean Reciprocal Rank (MRR)](https://en.wikipedi
 When the model has to pick the positive example out of a pool of 32, ranks the positive example highest most of the time.
 When the pool is significantly enlarged to 10.000 functions, it still ranks the positive example first or second in most cases.
 | Model   | Pool size | MRR  | Recall@1 |
 |---------|-----------|------|----------|
 | ASMBert | 32        | 0.99 | 0.99     |
 | ASMBert | 10.000    | 0.87 | 0.83     |
 ## Purpose and use of the model
 ### For which problem has the model been designed?
@@ -51,8 +48,6 @@ We do not see other applications for this model.
 ### To what problems is the model not applicable?
 This model has been finetuned on the semantic search task, for a generic ARM64-BERT model, please refer to the [other
 model](https://huggingface.co/NetherlandsForensicInstitute/ARM64bert) we have published.
 ## Data
 ### What data was used for training and evaluation?
@@ -64,7 +59,6 @@ in a maximum of 10 (5*2) different functions which are semantically similar i.e.
 The dataset is split into a train and a test set. This is done on project level, so all binaries and functions belonging to one project are part of
 either the train or the test set, not both. We have not performed any deduplication on the dataset for training.
 | set   | # functions |
 |-------|------------:|
 | train |  18,083,285 |
@@ -78,14 +72,12 @@ After training our models, we found out that something had gone wrong when compi
 the last line (instruction) of the previous function was included in the next. This has been fixed for the finetuning, but due to the long training process, and the
 good performance of the model despite the mistake, we have decided not to retrain the base model.
 ## Fairness Metrics
 ### Which metrics have been used to measure bias in the data/model and why?
 n.a.
-### What do those metrics show?
 n.a.
 ### Any other notable issues?

 library_name: sentence-transformers
 ---
 Model Card
+==========
 _Who to contact:_ fbda [at] nfi [dot] nl \
 TODO: add link to github repo
 although the typical Next Sentence Prediction has been replaced with Jump Target Prediction, as proposed in Wang et al.
 This architecture has subsequently been finetuned for semantic search purposes. We have followed the procedure proposed by [S-BERT](https://www.sbert.net/examples/applications/semantic-search/README.html).
 ### What is the output of the model?
 The model returns a vector of 768 dimensions for each function that it's given. These vectors can be compared to
 get an indication of which functions are similar to each other.
 When the model has to pick the positive example out of a pool of 32, ranks the positive example highest most of the time.
 When the pool is significantly enlarged to 10.000 functions, it still ranks the positive example first or second in most cases.
 | Model   | Pool size | MRR  | Recall@1 |
 |---------|-----------|------|----------|
 | ASMBert | 32        | 0.99 | 0.99     |
 | ASMBert | 10.000    | 0.87 | 0.83     |
 ## Purpose and use of the model
 ### For which problem has the model been designed?
 ### To what problems is the model not applicable?
 This model has been finetuned on the semantic search task, for a generic ARM64-BERT model, please refer to the [other
 model](https://huggingface.co/NetherlandsForensicInstitute/ARM64bert) we have published.
 ## Data
 ### What data was used for training and evaluation?
 The dataset is split into a train and a test set. This is done on project level, so all binaries and functions belonging to one project are part of
 either the train or the test set, not both. We have not performed any deduplication on the dataset for training.
 | set   | # functions |
 |-------|------------:|
 | train |  18,083,285 |
 the last line (instruction) of the previous function was included in the next. This has been fixed for the finetuning, but due to the long training process, and the
 good performance of the model despite the mistake, we have decided not to retrain the base model.
 ## Fairness Metrics
 ### Which metrics have been used to measure bias in the data/model and why?
 n.a.
+### What do those metrics show?
 n.a.
 ### Any other notable issues?