Skip to content
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

[GPTQ] fix tests #1598

Merged
merged 1 commit into from
Dec 14, 2023
Merged

[GPTQ] fix tests #1598

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
13 changes: 12 additions & 1 deletion tests/gptq/test_quantization.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -156,7 +156,10 @@ def test_serialization(self):
disable_exllama=self.disable_exllama,
exllama_config=self.exllama_config,
)
self.check_quantized_layers_type(quantized_model_from_saved, "cuda-old")
if self.disable_exllama:
self.check_quantized_layers_type(quantized_model_from_saved, "cuda-old")
else:
self.check_quantized_layers_type(quantized_model_from_saved, "exllama")

with torch.device("cuda"):
_ = AutoModelForCausalLM.from_pretrained(tmpdirname)
Expand All @@ -172,13 +175,15 @@ class GPTQTestExllama(GPTQTest):
EXPECTED_OUTPUTS.add("Hello my name is John, I am a professional photographer and I")
EXPECTED_OUTPUTS.add("Hello my name is jay and i am a student at university.")
EXPECTED_OUTPUTS.add("Hello my name is John, I am a student in the University of")
EXPECTED_OUTPUTS.add("Hello my name is Nate and I am a new member of the")


class GPTQTestActOrder(GPTQTest):
EXPECTED_OUTPUTS = set()
EXPECTED_OUTPUTS.add("Hello my name is jay and i am a student at university.")
EXPECTED_OUTPUTS.add("Hello my name is jessie and i am a very sweet and")
EXPECTED_OUTPUTS.add("Hello my name is nathalie, I am a young girl from")
EXPECTED_OUTPUTS.add("Hello my name is\nI am a student of the University of the'")

disable_exllama = True
desc_act = True
Expand Down Expand Up @@ -256,6 +261,11 @@ def test_exllama_max_input_length(self):
class GPTQTestExllamav2(GPTQTest):
desc_act = False
disable_exllama = True
EXPECTED_OUTPUTS = set()
EXPECTED_OUTPUTS.add("Hello my name is John, I am a professional photographer and I")
EXPECTED_OUTPUTS.add("Hello my name is jay and i am a student at university.")
EXPECTED_OUTPUTS.add("Hello my name is John, I am a student in the University of")
EXPECTED_OUTPUTS.add("Hello my name is Nate and I am a new member of the")

def test_generate_quality(self):
# don't need to test
Expand Down Expand Up @@ -300,6 +310,7 @@ class GPTQTestNoBlockCaching(GPTQTest):
EXPECTED_OUTPUTS.add("Hello my name is John, I am a professional photographer and I")
EXPECTED_OUTPUTS.add("Hello my name is jay and i am a student at university.")
EXPECTED_OUTPUTS.add("Hello my name is John, I am a student in the University of")
EXPECTED_OUTPUTS.add("Hello my name is Aiden and I am a very good looking")


class GPTQTestModuleQuant(GPTQTest):
Expand Down
Loading