Spaces:

codeparrot
/

code-generation-models

Runtime error

loubnabnl HF staff commited on May 25, 2022

Commit

9447150

•

1 Parent(s): 08d6a8d

update

Files changed (1) hide show

architectures/incoder.txt CHANGED Viewed

@@ -8,6 +8,8 @@
 [Causal Masking objective](https://arxiv.org/abs/2201.07520) is a hybrid approach of Causal and Masked language models, "it combines the benefit of per-token generation with optional bi-directionality specifically tailored to prompting".
 During the training of InCoder, spans of code were randomly masked and moved to the end of each file, which allows for bidirectional context. The figure below illustrates the training process:
-![image](https://huggingface.co/datasets/loubnabnl/repo-images/raw/main/incoder_small.png=x250)
 So in addition to program synthesis (via left-to-right generation), InCoder can also perform editing (via infilling). The model gives promising results in some zero-shot code infilling tasks such as type prediction, variable re-naming and comment generation.

 [Causal Masking objective](https://arxiv.org/abs/2201.07520) is a hybrid approach of Causal and Masked language models, "it combines the benefit of per-token generation with optional bi-directionality specifically tailored to prompting".
 During the training of InCoder, spans of code were randomly masked and moved to the end of each file, which allows for bidirectional context. The figure below illustrates the training process:
+![image](https://huggingface.co/datasets/loubnabnl/repo-images/raw/main/incoder_small.png)
+<img src="https://huggingface.co/datasets/loubnabnl/repo-images/raw/main/incoder.png" width="550" height="250" />
 So in addition to program synthesis (via left-to-right generation), InCoder can also perform editing (via infilling). The model gives promising results in some zero-shot code infilling tasks such as type prediction, variable re-naming and comment generation.