• File size measurements show the distribution of size of files.
• Files are classified in four categories based on their size (lines of code): 1-100 (very small files), 101-200 (small files), 201-500 (medium size files), 501-1000 (long files), 1001+(very long files).
• It is a good practice to keep files small. Long files may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.

• There are 34 files with 5,336 lines of code.
• 0 very long files (0 lines of code)
• 3 long files (1,827 lines of code)
• 7 medium size files (1,705 lines of codeclsfd_ftr_w_mp_ins)
• 6 small files (899 lines of code)
• 18 very small files (905 lines of code)

There are 20 files with lines longer than 120 characters. In total, there are 102 long lines.

File	# lines	# units	# long lines
HubApi.swift in Sources/Hub	608	22	38
Tokenizer.swift in Sources/Tokenizers	549	37	17
Generation.swift in Sources/Generation	82	2	5
Decoder.swift in Sources/Tokenizers	207	14	5
Downloader.swift in Sources/Hub	231	6	4
Package.swift in root	35	-	3
BertTokenizer.swift in Sources/Tokenizers	224	11	3
PreTokenizer.swift in Sources/Tokenizers	302	7	3
TokenLattice.swift in Sources/Tokenizers	101	6	3
Hub.swift in Sources/Hub	213	2	3
Config.swift in Sources/Hub	670	3	3
LanguageModel.swift in Sources/Models	174	1	2
LanguageModelTypes.swift in Sources/Models	26	4	2
Weights.swift in Sources/TensorUtils	77	1	2
BPETokenizer.swift in Sources/Tokenizers	165	9	2
PostProcessor.swift in Sources/Tokenizers	148	2	2
BinaryDistinct.swift in Sources/Hub	178	2	2
HubCLI.swift in Sources/HubCLI	80	3	1
GenerationConfig.swift in Sources/Generation	43	-	1
UnigramTokenizer.swift in Sources/Tokenizers	84	3	1