rename kernel files

Author: Qubitium

gptqmodel/looper/awq_processor.py

@@ -20,10 +20,10 @@
 from ..models._const import SUPPORTS_MODULE_TYPES
 from ..models.writer import (PROCESS_LOG_LAYER, PROCESS_LOG_MODULE, PROCESS_LOG_NAME,
                              PROCESS_LOG_TIME, PROCESS_USED_MEMORY, QUANT_LOG_LOSS, QUANT_LOG_NSAMPLES)
-from ..nn_modules.qlinear.awq_gemm import AwqGEMMQuantLinear
-from ..nn_modules.qlinear.awq_gemv import AwqGEMVQuantLinear
-from ..nn_modules.qlinear.awq_gemv_fast import AwqGEMVFastQuantLinear
-from ..nn_modules.qlinear.awq_marlin import AwqMarlinQuantLinear
+from ..nn_modules.qlinear.gemm_awq import AwqGEMMQuantLinear
+from ..nn_modules.qlinear.gemv_awq import AwqGEMVQuantLinear
+from ..nn_modules.qlinear.gemv_fast_awq import AwqGEMVFastQuantLinear
+from ..nn_modules.qlinear.marlin_awq import AwqMarlinQuantLinear
 from ..quantization.awq.modules.linear import WQLinear_GEMM, WQLinear_GEMV, WQLinear_GEMVFast, WQLinear_Marlin
 from ..quantization.awq.quantize.scale import apply_clip, apply_scale
 from ..quantization.awq.utils.module import append_str_prefix, get_op_name, get_op_by_name, set_op_by_name

gptqmodel/nn_modules/qlinear/awq_gemm.py renamed to gptqmodel/nn_modules/qlinear/gemm_awq.py

@@ -13,14 +13,14 @@
 
 from ..models._const import DEVICE, normalize_device
 from ..nn_modules.qlinear import BaseQuantLinear, PackableQuantLinear
-from ..nn_modules.qlinear.awq_exllama import AwqExllamaQuantLinear
-from ..nn_modules.qlinear.awq_exllamav2 import AwqExllamaV2QuantLinear
-from ..nn_modules.qlinear.awq_gemm import AwqGEMMQuantLinear
-from ..nn_modules.qlinear.awq_gemv import AwqGEMVQuantLinear
-from ..nn_modules.qlinear.awq_gemv_fast import AwqGEMVFastQuantLinear
-from ..nn_modules.qlinear.awq_machete import AwqMacheteQuantLinear
-from ..nn_modules.qlinear.awq_marlin import AwqMarlinQuantLinear
-from ..nn_modules.qlinear.awq_torch import AwqTorchQuantLinear
+from ..nn_modules.qlinear.exllama_awq import AwqExllamaQuantLinear
+from ..nn_modules.qlinear.exllamav2_awq import AwqExllamaV2QuantLinear
+from ..nn_modules.qlinear.gemm_awq import AwqGEMMQuantLinear
+from ..nn_modules.qlinear.gemv_awq import AwqGEMVQuantLinear
+from ..nn_modules.qlinear.gemv_fast_awq import AwqGEMVFastQuantLinear
+from ..nn_modules.qlinear.machete_awq import AwqMacheteQuantLinear
+from ..nn_modules.qlinear.marlin_awq import AwqMarlinQuantLinear
+from ..nn_modules.qlinear.torch_awq import AwqTorchQuantLinear
 from ..nn_modules.qlinear.bitblas import BitBLASQuantLinear
 from ..nn_modules.qlinear.exllama import ExllamaQuantLinear
 from ..nn_modules.qlinear.exllama_eora import ExllamaEoraQuantLinear

gptqmodel/nn_modules/qlinear/awq_gemv.py renamed to gptqmodel/nn_modules/qlinear/gemv_awq.py

@@ -46,7 +46,7 @@
     SUPPORTS_MODULE_TYPES,
 )
 from ..nn_modules.qlinear import BaseQuantLinear
-from ..nn_modules.qlinear.awq_exllamav2 import AwqExllamaV2QuantLinear
+from ..nn_modules.qlinear.exllamav2_awq import AwqExllamaV2QuantLinear
 from ..nn_modules.qlinear.exllama import ExllamaQuantLinear
 from ..nn_modules.qlinear.exllamav2 import ExllamaV2QuantLinear
 from ..quantization import FORMAT, QuantizeConfig

gptqmodel/nn_modules/qlinear/awq_gemv_fast.py renamed to gptqmodel/nn_modules/qlinear/gemv_fast_awq.py

@@ -15,11 +15,11 @@
 from parameterized import parameterized
 from transformers import AutoTokenizer
 
-from gptqmodel.nn_modules.qlinear.awq_gemm import AwqGEMMQuantLinear
-from gptqmodel.nn_modules.qlinear.awq_gemv import AwqGEMVQuantLinear
-from gptqmodel.nn_modules.qlinear.awq_gemv_fast import AwqGEMVFastQuantLinear
-from gptqmodel.nn_modules.qlinear.awq_machete import AwqMacheteQuantLinear
-from gptqmodel.nn_modules.qlinear.awq_marlin import AwqMarlinQuantLinear
+from gptqmodel.nn_modules.qlinear.gemm_awq import AwqGEMMQuantLinear
+from gptqmodel.nn_modules.qlinear.gemv_awq import AwqGEMVQuantLinear
+from gptqmodel.nn_modules.qlinear.gemv_fast_awq import AwqGEMVFastQuantLinear
+from gptqmodel.nn_modules.qlinear.machete_awq import AwqMacheteQuantLinear
+from gptqmodel.nn_modules.qlinear.marlin_awq import AwqMarlinQuantLinear
 from gptqmodel.quantization import FORMAT, METHOD, QUANT_CONFIG_FILENAME
 from gptqmodel.utils.machete import _validate_machete_device_support, machete_import_exception