bert-base-cased · azureml v18 ↗
- source model id
- azureml/bert-base-cased@18
- hosting
- foundry-managed-compute
- platforms
- Microsoft Foundry
- license
- apache-2.0
- architecture
- input_modalities
- output_modalities
- capabilities
BERT is a transformers model pretrained on a large corpus of English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling them in any way (w…
azureFoundry/azureml/bert-base-cased@18
BERT is a transformers model pretrained on a large corpus of English data in a self-supervised fashion. This means it was pretrained on the raw texts only, with no humans labelling them in any way (which is why it can use lots of publicly available data) with an automatic process to generate inputs and labels from those texts. More precisely, it was pretrained with two objectives:
This way, the model learns an inner representation of the English language that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled sentences for instance, you can train a standard classifier using the features produced by the BERT model as inputs.
It was introduced in this paper and first released in this repository. This model is case-sensitive: it makes a difference between english and English.
The BERT model was pretrained on BookCorpus, a dataset consisting of 11,038 unpublished books and English Wikipedia (excluding lists, tables and headers).
The texts are tokenized using WordPiece and a vocabulary size of 30,000. The inputs of the model are then of the form:
[CLS] Sentence A [SEP] Sentence B [SEP]
With probability 0.5, sentence A and sentence B correspond to two consecutive sentences in the original corpus and in the other cases, it's another random sentence in the corpus. Note that what is considered a sentence here is a consecutive span of text usually longer than a single sentence. The only constrain is that the result with the two "sentences" has a combined length of less than 512 tokens.
The details of the masking procedure for each sentence are the following:
[MASK].The model was trained on 4 cloud TPUs in Pod configuration (16 TPU chips total) for one million steps with a batch size of 256. The sequence length was limited to 128 tokens for 90% of the steps and 512 for the remaining 10%. The optimizer used is Adam with a learning rate of 1e-4, \(\beta_{1} = 0.9\) and \(\beta_{2} = 0.999\), a weight decay of 0.01, learning rate warmup for 10,000 steps and linear decay of the learning rate after.
When fine-tuned on downstream tasks, this model achieves the following results:
Glue test results:
| Task | MNLI-(m/mm) | QQP | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE | Average |
|---|---|---|---|---|---|---|---|---|---|
| 84.6/83.4 | 71.2 | 90.5 | 93.5 | 52.1 | 85.8 | 88.9 | 66.4 | 79.6 |
Even if the training data used for this model could be characterized as fairly neutral, this model can have biased predictions. This bias will also affect all fine-tuned versions of this model.
You can use the raw model for either masked language modeling or next sentence prediction, but it's mostly intended to be fine-tuned on a downstream task.
Note that this model is primarily aimed at being fine-tuned on tasks that use the whole sentence (potentially masked) to make decisions, such as sequence classification, token classification or question answering. For tasks such as text generation you should look at model like GPT2.
| Task | Use case | Dataset | Python sample (Notebook) | CLI with YAML |
|---|---|---|---|---|
| Fill Mask | Fill Mask | rcds/wikipedia-for-mask-filling | evaluate-model-fill-mask.ipynb | evaluate-model-fill-mask.yml |
| Inference type | Python sample (Notebook) |
|---|---|
| Real time | sdk-example.ipynb |
| Real time | fill-mask-online-endpoint.ipynb |
{
"input_data": [
"Paris is the [MASK] of France.",
"Today is a [MASK] day!"
]
}
[
"capital",
"beautiful"
]
1–1 / 1
1–3 / 3
Microsoft Foundry · Model page ↗
1–1 / 1
These charts contain model-linked results. The charts above also include unlinked source profiles. Compare scores within each evaluation. Model quality, usage, transparency and system performance measure different things.