huggingface / optimum-quanto Duplication

Intro

For duplication, we look at places in code where there are 6 or more lines of code that are exactly the same.
Before duplication is calculated, the code is cleaned to remove empty lines, comments, and frequently duplicated constructs such as imports.
You should aim at having as little as possible (<5%) of duplicated code as high-level of duplication can lead to maintenance difficulties, poor factoring, and logical contradictions.

Learn more...

Duplication Overall

19% duplication:

8,156 cleaned lines of cleaned code (without empty lines, comments, and frequently duplicated constructs such as imports)
1,587 duplicated lines

116 duplicates

dependency graphs: 2D graph | 3D graph | 3D graph (with duplicates)...

Duplication per Extension

Duplication per Component (primary)

Longest Duplicates

The list of 50 longest duplicates.

See data for all 116 duplicates...

Size	#	Folders	Files	Lines	Code
77	x 2	optimum/quanto/library/extensions/cuda optimum/quanto/library/extensions/hip	unpack.cu unpack.cu	1:97 (100%) 1:97 (100%)	view
26	x 2	bench/generation/metrics bench/kernels	latency.py benchmark.py	25:55 (32%) 37:67 (32%)	view
24	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu fp8_marlin.cu	42:65 (2%) 252:275 (2%)	view
23	x 2	optimum/quanto/tensor/weights/awq optimum/quanto/tensor/weights/marlin/int4	qbits.py qbits.py	112:138 (23%) 117:143 (22%)	view
20	x 2	optimum/quanto/tensor/weights/awq optimum/quanto/tensor/weights	qbits.py qbits.py	116:138 (20%) 222:244 (10%)	view
20	x 2	optimum/quanto/tensor/weights optimum/quanto/tensor/weights	qbits.py qbytes.py	266:290 (10%) 268:292 (9%)	view
20	x 2	optimum/quanto/tensor/weights/marlin/int4 optimum/quanto/tensor/weights	qbits.py qbits.py	121:143 (19%) 222:244 (10%)	view
18	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	135:152 (1%) 590:607 (1%)	view
18	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	169:187 (1%) 624:642 (1%)	view
17	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	217:234 (1%) 670:687 (1%)	view
17	x 2	optimum/quanto/tensor/weights/marlin/int4 optimum/quanto/tensor/weights/tinygemm	packed.py packed.py	115:133 (16%) 97:117 (21%)	view
15	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	marlin_cuda_kernel.cu marlin_cuda_kernel.cuh	1:15 (1%) 1:15 (44%)	view
14	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	421:436 (1%) 862:876 (1%)	view
14	x 2	optimum/quanto/tensor/weights/marlin/fp8 optimum/quanto/tensor/weights/marlin/int4	packed.py packed.py	223:239 (10%) 119:133 (13%)	view
14	x 2	optimum/quanto/tensor/weights/marlin/fp8 optimum/quanto/tensor/weights/tinygemm	packed.py packed.py	223:239 (10%) 101:117 (17%)	view
14	x 2	bench/generation bench/generation/setup	evaluate_model.py quanto.py	35:48 (14%) 24:37 (28%)	view
13	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu marlin_cuda_kernel.cuh	4:16 (1%) 3:15 (38%)	view
13	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu marlin_cuda_kernel.cu	4:16 (1%) 3:15 (1%)	view
13	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu marlin_cuda_kernel.cu	229:242 (1%) 170:183 (1%)	view
13	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	154:167 (1%) 609:622 (1%)	view
13	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	189:202 (1%) 644:657 (1%)	view
13	x 2	bench/generation bench/generation	evaluate_configurations.py evaluate_model.py	79:93 (16%) 117:131 (13%)	view
12	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	278:290 (1%) 731:742 (1%)	view
12	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	254:266 (1%) 707:719 (1%)	view
11	x 2	optimum/quanto/models optimum/quanto/models	diffusers_models.py transformers_models.py	51:65 (11%) 45:59 (11%)	view
11	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	marlin_cuda_kernel.cu marlin_cuda_kernel.cuh	761:771 (1%) 24:34 (32%)	view
11	x 2	bench/generation bench/kernels	evaluate_configurations.py benchmark.py	83:93 (13%) 101:111 (13%)	view
11	x 2	bench/generation bench/kernels	evaluate_model.py benchmark.py	121:131 (11%) 101:111 (13%)	view
10	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu marlin_cuda_kernel.cu	348:357 (<1%) 274:283 (1%)	view
10	x 2	optimum/quanto/nn optimum/quanto/nn	qconv2d.py qlinear.py	28:37 (33%) 28:37 (40%)	view
10	x 2	optimum/quanto/tensor/weights optimum/quanto/tensor/weights/tinygemm	qbits.py qbits.py	225:235 (5%) 133:143 (9%)	view
10	x 2	optimum/quanto/tensor/weights optimum/quanto/tensor/weights	qbits.py qbytes.py	168:179 (5%) 173:184 (4%)	view
10	x 2	optimum/quanto/tensor/weights/awq optimum/quanto/tensor/weights/tinygemm	qbits.py qbits.py	119:129 (10%) 133:143 (9%)	view
10	x 2	optimum/quanto/tensor/weights/marlin/int4 optimum/quanto/tensor/weights/tinygemm	qbits.py qbits.py	124:134 (9%) 133:143 (9%)	view
10	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	454:463 (1%) 893:902 (1%)	view
9	x 2	optimum/quanto/models optimum/quanto/models	diffusers_models.py transformers_models.py	127:137 (9%) 124:134 (9%)	view
9	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	524:533 (<1%) 549:557 (<1%)	view
9	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	368:377 (<1%) 809:818 (<1%)	view
9	x 2	bench/generation bench/generation	evaluate_configurations.py evaluate_model.py	66:74 (11%) 88:96 (9%)	view
8	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	988:995 (<1%) 1001:1008 (<1%)	view
8	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/marlin	gemv_cuda.cu fp8_marlin.cu	10:18 (2%) 10:18 (<1%)	view
8	x 2	optimum/quanto/tensor/weights optimum/quanto/tensor/weights	qbytes.py qbytes.py	44:51 (3%) 53:60 (3%)	view
8	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	948:955 (<1%) 961:969 (<1%)	view
8	x 2	optimum/quanto/library/extensions/cuda/awq optimum/quanto/library/extensions/cuda/awq/v2	dequantize.cuh gemv_cuda.cu	4:11 (11%) 19:27 (2%)	view
8	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu marlin_cuda_kernel.cu	478:486 (<1%) 367:375 (1%)	view
8	x 2	optimum/quanto/library/extensions/cuda/marlin optimum/quanto/library/extensions/cuda/marlin	fp8_marlin.cu marlin_cuda_kernel.cu	804:812 (<1%) 628:636 (1%)	view
8	x 2	optimum/quanto/library/extensions/cuda/awq/v2 optimum/quanto/library/extensions/cuda/awq/v2	gemm_cuda.cu gemm_cuda.cu	327:334 (<1%) 773:780 (<1%)	view
8	x 2	optimum/quanto/tensor/weights/awq optimum/quanto/tensor/weights/marlin/int4	qbits.py qbits.py	37:46 (8%) 41:50 (7%)	view
7	x 2	optimum/quanto/library/extensions/cuda optimum/quanto/library/extensions/hip	unpack.cu unpack.cu	75:83 (9%) 48:56 (9%)	view
7	x 2	optimum/quanto/library/extensions/cuda optimum/quanto/library/extensions/cuda	unpack.cu unpack.cu	48:56 (9%) 75:83 (9%)	view