Skip to content

FIX Multi GPU tests: explicit device map #2484

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
BenjaminBossan merged 2 commits into from
Apr 11, 2025
Merged

FIX Multi GPU tests: explicit device map #2484

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
efb0182
FIX Multi GPU tests: explicit device map
BenjaminBossan Apr 8, 2025
344a1a7
Reviewer feedback: better device_map
BenjaminBossan Apr 8, 2025
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
93 changes: 89 additions & 4 deletions tests/test_gpu_examples.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1649,16 +1649,37 @@ def test_causal_lm_training_multi_gpu(self):
Test the CausalLM training on a multi-GPU device. The test would simply fail if the adapters are not set
correctly.
"""
device_map = {
"model.decoder.embed_tokens": 0,
"lm_head": 0,
"model.decoder.embed_positions": 0,
"model.decoder.project_out": 0,
"model.decoder.project_in": 0,
"model.decoder.layers.0": 0,
"model.decoder.layers.1": 0,
"model.decoder.layers.2": 0,
"model.decoder.layers.3": 0,
"model.decoder.layers.4": 0,
"model.decoder.layers.5": 0,
"model.decoder.layers.6": 1,
"model.decoder.layers.7": 1,
"model.decoder.layers.8": 1,
"model.decoder.layers.9": 1,
"model.decoder.layers.10": 1,
"model.decoder.layers.11": 1,
"model.decoder.final_layer_norm": 1,
}

with tempfile.TemporaryDirectory() as tmp_dir:
model = AutoModelForCausalLM.from_pretrained(
self.causal_lm_model_id,
torch_dtype=torch.float16,
device_map="auto",
device_map=device_map,
quantization_config=self.quantization_config,
)

assert set(model.hf_device_map.values()) == set(range(device_count))
assert {p.device.index for p in model.parameters()} == set(range(device_count))

model = prepare_model_for_kbit_training(model)

Expand Down Expand Up @@ -3182,14 +3203,35 @@ def test_causal_lm_training_multi_gpu(self):
Test the CausalLM training on a multi-GPU device. The test would simply fail if the adapters are not set
correctly.
"""
device_map = {
"model.decoder.embed_tokens": 0,
"lm_head": 0,
"model.decoder.embed_positions": 0,
"model.decoder.project_out": 0,
"model.decoder.project_in": 0,
"model.decoder.layers.0": 0,
"model.decoder.layers.1": 0,
"model.decoder.layers.2": 0,
"model.decoder.layers.3": 0,
"model.decoder.layers.4": 0,
"model.decoder.layers.5": 0,
"model.decoder.layers.6": 1,
"model.decoder.layers.7": 1,
"model.decoder.layers.8": 1,
"model.decoder.layers.9": 1,
"model.decoder.layers.10": 1,
"model.decoder.layers.11": 1,
"model.decoder.final_layer_norm": 1,
}

with tempfile.TemporaryDirectory() as tmp_dir:
model = AutoModelForCausalLM.from_pretrained(
self.causal_lm_model_id,
device_map="auto",
device_map=device_map,
)

assert set(model.hf_device_map.values()) == set(range(device_count))
assert {p.device.index for p in model.parameters()} == set(range(device_count))

model = prepare_model_for_kbit_training(model)

Expand Down Expand Up @@ -3579,16 +3621,38 @@ def test_causal_lm_training_single_gpu_torchao_int4_raises(self):
def test_causal_lm_training_multi_gpu_torchao(self, quant_type):
from transformers import TorchAoConfig

device_map = {
"model.decoder.embed_tokens": 0,
"lm_head": 0,
"model.decoder.embed_positions": 0,
"model.decoder.project_out": 0,
"model.decoder.project_in": 0,
"model.decoder.layers.0": 0,
"model.decoder.layers.1": 0,
"model.decoder.layers.2": 0,
"model.decoder.layers.3": 0,
"model.decoder.layers.4": 0,
"model.decoder.layers.5": 0,
"model.decoder.layers.6": 1,
"model.decoder.layers.7": 1,
"model.decoder.layers.8": 1,
"model.decoder.layers.9": 1,
"model.decoder.layers.10": 1,
"model.decoder.layers.11": 1,
"model.decoder.final_layer_norm": 1,
}

with tempfile.TemporaryDirectory() as tmp_dir:
quantization_config = TorchAoConfig(quant_type=quant_type)
model = AutoModelForCausalLM.from_pretrained(
self.causal_lm_model_id,
device_map="auto",
device_map=device_map,
quantization_config=quantization_config,
torch_dtype=torch.bfloat16,
)

assert set(model.hf_device_map.values()) == set(range(device_count))
assert {p.device.index for p in model.parameters()} == set(range(device_count))

model = prepare_model_for_kbit_training(model)
model.model_parallel = True
Expand Down Expand Up @@ -3640,15 +3704,36 @@ def test_causal_lm_training_multi_gpu_torchao_int4_raises(self):
# TODO: Once proper torchao support for int4 is added, remove this test and add int4 to supported_quant_types
from transformers import TorchAoConfig

device_map = {
"model.decoder.embed_tokens": 0,
"lm_head": 0,
"model.decoder.embed_positions": 0,
"model.decoder.project_out": 0,
"model.decoder.project_in": 0,
"model.decoder.layers.0": 0,
"model.decoder.layers.1": 0,
"model.decoder.layers.2": 0,
"model.decoder.layers.3": 0,
"model.decoder.layers.4": 0,
"model.decoder.layers.5": 0,
"model.decoder.layers.6": 1,
"model.decoder.layers.7": 1,
"model.decoder.layers.8": 1,
"model.decoder.layers.9": 1,
"model.decoder.layers.10": 1,
"model.decoder.layers.11": 1,
"model.decoder.final_layer_norm": 1,
}
quantization_config = TorchAoConfig(quant_type="int4_weight_only")
model = AutoModelForCausalLM.from_pretrained(
self.causal_lm_model_id,
device_map="auto",
device_map=device_map,
quantization_config=quantization_config,
torch_dtype=torch.bfloat16,
)

assert set(model.hf_device_map.values()) == set(range(device_count))
assert {p.device.index for p in model.parameters()} == set(range(device_count))

model = prepare_model_for_kbit_training(model)
model.model_parallel = True
Expand Down
Loading