Statements (103)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:Model
gptkb:Flight |
| gptkbp:application |
bug detection
documentation generation automated code review code translation |
| gptkbp:architecture |
gptkb:Transformers
|
| gptkbp:available_on |
gptkb:Hugging_Face_Model_Hub
|
| gptkbp:based_on |
gptkb:BERT
BERT architecture |
| gptkbp:community_support |
active
|
| gptkbp:designed_for |
gptkb:C_programming_language
natural languages |
| gptkbp:developed_by |
gptkb:Microsoft
gptkb:Microsoft_Research |
| gptkbp:evaluates |
F1 score
accuracy |
| gptkbp:feature |
efficiency
extensibility flexibility interoperability modularity reusability scalability robustness transfer learning contextual embeddings adaptability customizability multi-modal capabilities support for various frameworks support for various languages support for various platforms language-agnostic fine-tuning capabilities user-friendly API |
| gptkbp:has_achieved |
state-of-the-art results
|
| gptkbp:has_applications_in |
gptkb:Artificial_Intelligence
gptkb:machine_learning gptkb:Software_Development |
| gptkbp:has_features |
contextual embeddings
multi-modal learning pre-training and fine-tuning cross-lingual capabilities tokenization for code |
| gptkbp:has_limitations |
computationally intensive
requires large datasets depends on quality of training data may not generalize well to unseen languages |
| gptkbp:has_publications |
Code BERT: A Pre-trained Model for Programming and Natural Languages
|
| https://www.w3.org/2000/01/rdf-schema#label |
Code BERT
|
| gptkbp:improves |
programming tasks
natural language tasks |
| gptkbp:input_output |
tokenized text
text and code |
| gptkbp:is_cited_in |
over 1000
|
| gptkbp:is_compared_to |
gptkb:GPT-3
gptkb:BERT gptkb:Ro_BERTa |
| gptkbp:is_evaluated_by |
gptkb:Code_Search_Net_dataset
gptkb:GLUE_benchmark |
| gptkbp:is_part_of |
AI for Code
|
| gptkbp:is_trained_in |
natural language text
Git Hub repositories Git Hub code repositories |
| gptkbp:is_used_in |
code refactoring
bug detection automated code review educational tools for programming intelligent code editors |
| gptkbp:language |
gptkb:C_programming_language
|
| gptkbp:license |
MIT License
|
| gptkbp:performance |
state-of-the-art
|
| gptkbp:provides_information_on |
natural language and code pairs
|
| gptkbp:published_in |
gptkb:2020
|
| gptkbp:related_to |
gptkb:Natural_Language_Processing
gptkb:developers gptkb:Artificial_Intelligence gptkb:machine_learning gptkb:Deep_Learning |
| gptkbp:release_year |
gptkb:2020
|
| gptkbp:supports |
gptkb:Ruby
gptkb:Java gptkb:PHP gptkb:Python gptkb:C# gptkb:Go gptkb:Swift gptkb:Java_Script code completion multiple programming languages code search code summarization code translation |
| gptkbp:training |
supervised learning
unsupervised learning |
| gptkbp:tuning |
natural language processing tasks
|
| gptkbp:used_for |
code completion
code search code summarization |
| gptkbp:uses |
gptkb:Transformers
|
| gptkbp:bfsParent |
gptkb:Super_GLUE_benchmark
|
| gptkbp:bfsLayer |
5
|