fix ut & add torch2.0 for ci

.github/workflows/build.yml

@@ -29,7 +29,7 @@ jobs:
     strategy:
       matrix:
         python-version: [3.7]
-        torch: [1.8.0, 1.9.0, 1.10.0, 1.11.0, 1.12.0, 1.13.0]
+        torch: [1.8.0, 1.9.0, 1.10.0, 1.11.0, 1.12.0, 1.13.0, 2.0.0]
         include:
           - torch: 1.8.0
             torch_version: 1.8
@@ -73,6 +73,10 @@ jobs:
             torch_version: 1.13
             torchvision: 0.14.0
             python-version: 3.8
+          - torch: 2.0.0
+            torch_version: 2.0.0
+            torchvision: 0.15.0
+            python-version: 3.8
 
     steps:
       - uses: actions/checkout@v2

mmrazor/implementations/quantization/gptq/__init__.py

@@ -1,16 +1,11 @@
 # Copyright (c) OpenMMLab. All rights reserved.
 from .compressor import GPTQCompressor
-from .custom_autotune import (Autotuner, autotune,
-                              matmul248_kernel_config_pruner)
 from .gptq import GPTQMixIn
 from .ops import GPTQConv2d, GPTQLinear, TritonGPTQLinear
 from .quantizer import Quantizer
 
 __all__ = [
     'GPTQCompressor',
-    'Autotuner',
-    'autotune',
-    'matmul248_kernel_config_pruner',
     'GPTQMixIn',
     'GPTQConv2d',
     'GPTQLinear',

mmrazor/implementations/quantization/gptq/custom_autotune.py

@@ -12,7 +12,7 @@ try:
     import triton
 except ImportError:
     from mmrazor.utils import get_package_placeholder
-    triton = get_package_placeholder('please install triton with pip')
+    triton = get_package_placeholder('triton >= 2.0.0')
 
 
 class Autotuner(triton.KernelInterface):

mmrazor/implementations/quantization/gptq/ops.py

@@ -350,7 +350,7 @@ try:
         c_mask = (offs_am[:, None] < M) & (offs_bk[None, :] < K)
         tl.store(c_ptrs, accumulator, mask=c_mask)
 except:  # noqa: E722
-    print('trioton not installed.')
+    print('triton not installed.')
 
 
 def matmul248(input, qweight, scales, qzeros, g_idx, bits, maxq):

requirements/tests.txt

@@ -7,6 +7,6 @@ nbformat
 numpy < 1.24.0 # A temporary solution for tests with mmdet.
 onnx
 pytest
-triton
+triton==2.0.0
 xdoctest >= 0.10.0
 yapf