Table 3: Dataset statistics. Java Python All - Size 26 GB19 GB All - Nb files 7.9M 3.6M Av. nb of tokens / file 718 1245 Av. nb of identifiers / file 25.9 41.8 Input Code Proposed Function Name

DOBF finds relevant names for Java methods without copying any part of the other tokens. For instance, in the first example, it understands that this function is used to get environment variables. Although l and a are not very informative, they indicate that the variable is a list or an array. These three functions perform graph traversals. Linear(input_size, ( 4 * hidden_size), bias = bias) self .