Skip to content

[FIX] test_packable.py #1357

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 1 commit into from
Feb 27, 2025
Merged

[FIX] test_packable.py #1357

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
37 changes: 28 additions & 9 deletions tests/test_packable.py
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,14 +1,14 @@
import os.path
import unittest
from typing import Dict

import torch
from parameterized import parameterized
from safetensors.torch import load_file

from gptqmodel import BACKEND, GPTQModel
from gptqmodel.utils.model import find_modules
from gptqmodel.utils.model import find_modules, convert_gptq_v2_to_v1_format

from gptqmodel.nn_modules.qlinear.bitblas import BitBLASQuantLinear # noqa: E402
from gptqmodel.nn_modules.qlinear.dynamic_cuda import DynamicCudaQuantLinear # noqa: E402
from gptqmodel.nn_modules.qlinear.exllama import ExllamaQuantLinear # noqa: E402
from gptqmodel.nn_modules.qlinear.exllamav2 import ExllamaV2QuantLinear # noqa: E402
Expand Down Expand Up @@ -45,18 +45,37 @@ def setUpClass(cls):
del weights

@parameterized.expand(
list(QLINEAR_DICT.keys())
[
(BACKEND.EXLLAMA_V1, {"qweight": True, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.EXLLAMA_V2, {"qweight": False, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.TRITON, {"qweight": True, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.CUDA, {"qweight": True, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.TORCH, {"qweight": True, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.CUDA, {"qweight": True, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.IPEX, {"qweight": True, "qzeros": True, "scales": True, "g_idx": True}),
(BACKEND.MARLIN, {"qweight": False, "qzeros": True, "scales": False, "g_idx": False}),
(BACKEND.MARLIN_FP16, {"qweight": False, "qzeros": True, "scales": False, "g_idx": False}),
]
)
def test_post_init(self, backend: BACKEND):
def test_post_init(self, backend: BACKEND, equal: Dict[str, bool]):
model = GPTQModel.load(self.model_id, backend=backend, device_map="auto")
model = convert_gptq_v2_to_v1_format(model, model.quantize_config, self.QLINEAR_DICT[backend])

module = find_modules(model.model, [self.QLINEAR_DICT[backend]])[self.TARGET]
# assert len(modules) == 1
state_dict = module.state_dict()
device = module.qweight.data.device
assert torch.equal(state_dict["qweight"], self.orgi_module_qweight.to(device))
# assert torch.equal(state_dict["qzeros"], self.orgi_module_qzeros.to(device))
assert torch.equal(state_dict["scales"], self.orgi_module_scales.to(device))
assert torch.equal(state_dict["g_idx"], self.orgi_module_g_idx.to(device))

def check(stat, key, expect_tensor):
r = torch.equal(stat[key], expect_tensor)
if equal[key]:
assert r
else:
assert not r

check(state_dict, "qweight", self.orgi_module_qweight.to(device))
check(state_dict, "qzeros", self.orgi_module_qzeros.to(device))
check(state_dict, "scales", self.orgi_module_scales.to(device))
check(state_dict, "g_idx", self.orgi_module_g_idx.to(device))

del state_dict
del module
Expand Down