mirror of
https://github.com/NVIDIA/Model-Optimizer.git
synced 2026-10-02 03:14:52 +08:00
Keval MorabiaandRinZ27
5dc17dfd15
### What does this PR do? - Add secure checkpoint loading support using `torch.serialization.add_safe_globals([cls])`. This also removes 1 existing pickle usage. - Remove hard-coded `trust_remote_code=True` - Replaces https://github.com/NVIDIA/Model-Optimizer/pull/1056 by @RinZ27 ### Testing <!-- Mention how have you tested your change if applicable. --> CICD tests ran ### Before your PR is "*Ready for review*" Make sure you read and follow [Contributor guidelines](https://github.com/NVIDIA/Model-Optimizer/blob/main/CONTRIBUTING.md) and your commits are signed (`git commit -s -S`). Make sure you read and follow the [Security Best Practices](https://github.com/NVIDIA/Model-Optimizer/blob/main/SECURITY.md#security-coding-practices-for-contributors) (e.g. avoiding hardcoded `trust_remote_code=True`, `torch.load(..., weights_only=False)`, `pickle`, etc.). - Is this change backward compatible?: ✅ <!--- If ❌, explain why. --> - If you copied code from any other sources or added a new PIP dependency, did you follow guidance in `CONTRIBUTING.md`: ✅ <!--- Mandatory --> - Did you write any new necessary tests?: ✅ <!--- Mandatory for new features or examples. --> - Did you update [Changelog](https://github.com/NVIDIA/Model-Optimizer/blob/main/CHANGELOG.rst)?: ✅ <!--- Only for new features, API changes, critical bug fixes or backward incompatible changes. --> ### Additional Information NVBug: 5999336 <!-- This is an auto-generated comment: release notes by coderabbit.ai --> ## Summary by CodeRabbit * **New Features** * Added safe checkpoint save/load helpers and a --trust_remote_code CLI flag in examples to control remote-code loading. * **Bug Fixes** * Checkpoint loading now defaults to safer, weights-only semantics to reduce arbitrary-code exposure. * **Documentation** * CHANGELOG updated with security guidance and opt-in procedure for unsafe checkpoint loading. * **Tests** * New unit tests validating the safe-load behavior. <!-- end of auto-generated comment: release notes by coderabbit.ai --> --------- Signed-off-by: RinZ27 <222222878+RinZ27@users.noreply.github.com> Signed-off-by: Keval Morabia <28916987+kevalmorabia97@users.noreply.github.com> Co-authored-by: RinZ27 <222222878+RinZ27@users.noreply.github.com>
348 lines
13 KiB
Python
348 lines
13 KiB
Python
# SPDX-FileCopyrightText: Copyright (c) 2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
|
|
# SPDX-License-Identifier: Apache-2.0
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# http://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
|
|
import torch
|
|
import torch.nn as nn
|
|
from fire import Fire
|
|
from peft import PeftModel
|
|
from pydantic import BaseModel, ConfigDict
|
|
from transformers import (
|
|
AutoModelForCausalLM,
|
|
AutoModelForSeq2SeqLM,
|
|
AutoTokenizer,
|
|
PreTrainedModel,
|
|
PreTrainedTokenizer,
|
|
)
|
|
|
|
try:
|
|
import tensorrt_llm
|
|
from tensorrt_llm.runtime import PYTHON_BINDINGS, ModelRunner
|
|
|
|
if PYTHON_BINDINGS:
|
|
from tensorrt_llm.runtime import ModelRunnerCpp
|
|
except ImportError:
|
|
tensorrt_llm = None
|
|
PYTHON_BINDINGS = None
|
|
ModelRunner = None
|
|
ModelRunnerCpp = None
|
|
|
|
try:
|
|
from awq import AutoAWQForCausalLM
|
|
except ImportError:
|
|
AutoAWQForCausalLM = None
|
|
|
|
|
|
class EvalModel(BaseModel):
|
|
model_config = ConfigDict(arbitrary_types_allowed=True)
|
|
|
|
model_path: str
|
|
trust_remote_code: bool = False
|
|
max_input_length: int = 512
|
|
max_output_length: int = 512
|
|
dtype: str = "auto"
|
|
|
|
def run(self, prompt: str, **kwargs) -> str:
|
|
raise NotImplementedError
|
|
|
|
def count_text_length(self, text: str) -> int:
|
|
raise NotImplementedError
|
|
|
|
def check_valid_length(self, text: str) -> bool:
|
|
return self.count_text_length(text) <= self.max_input_length
|
|
|
|
def load(self):
|
|
raise NotImplementedError
|
|
|
|
|
|
class SeqToSeqModel(EvalModel):
|
|
model: PreTrainedModel | None = None
|
|
tokenizer: PreTrainedTokenizer | None = None
|
|
lora_path: str = ""
|
|
device: str = "cuda"
|
|
load_8bit: bool = False
|
|
|
|
def load(self):
|
|
if self.model is None:
|
|
args = {}
|
|
if self.device == "cuda":
|
|
args.update(device_map="auto")
|
|
if self.load_8bit:
|
|
args.update(device_map="auto", load_in_8bit=True)
|
|
if self.dtype != "auto":
|
|
args.update(torch_dtype=getattr(torch, self.dtype))
|
|
else:
|
|
args.update(torch_dtype="auto")
|
|
self.model = AutoModelForSeq2SeqLM.from_pretrained(
|
|
self.model_path, trust_remote_code=self.trust_remote_code, **args
|
|
)
|
|
print_gpu_utilization()
|
|
if self.lora_path:
|
|
self.model = PeftModel.from_pretrained(self.model, self.lora_path)
|
|
self.model.eval()
|
|
if "device_map" not in args:
|
|
self.model.to(self.device)
|
|
if self.tokenizer is None:
|
|
self.tokenizer = AutoTokenizer.from_pretrained(
|
|
self.model_path, trust_remote_code=self.trust_remote_code
|
|
)
|
|
|
|
def run(self, prompt: str, **kwargs) -> str:
|
|
self.load()
|
|
assert self.model is not None, "Model must be loaded before running."
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded before running."
|
|
device = self.model.device if hasattr(self.model, "device") else self.device
|
|
inputs = self.tokenizer(prompt, return_tensors="pt").to(device)
|
|
outputs = self.model.generate(
|
|
**inputs,
|
|
max_length=self.max_output_length,
|
|
**kwargs,
|
|
)
|
|
return self.tokenizer.decode(outputs[0], skip_special_tokens=True)
|
|
|
|
def count_text_length(self, text: str) -> int:
|
|
self.load()
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded to count text length."
|
|
return len(self.tokenizer(text).input_ids)
|
|
|
|
def get_choice(self, text: str, **kwargs) -> tuple[float, float]:
|
|
self.load()
|
|
assert self.model is not None, "Model must be loaded to get choices."
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded to get choices."
|
|
inputs = self.tokenizer(text, return_tensors="pt").to(self.device)
|
|
start_token = torch.tensor([[self.tokenizer.pad_token_id]], dtype=torch.long).to(
|
|
self.device
|
|
)
|
|
with torch.no_grad():
|
|
predictions = self.model(
|
|
**inputs,
|
|
decoder_input_ids=start_token,
|
|
**kwargs,
|
|
).logits[0, 0]
|
|
a_index = self.tokenizer("A", add_special_tokens=False).input_ids[0]
|
|
b_index = self.tokenizer("B", add_special_tokens=False).input_ids[0]
|
|
a = float(predictions[a_index].cpu())
|
|
b = float(predictions[b_index].cpu())
|
|
return a, b
|
|
|
|
|
|
class CausalModel(SeqToSeqModel):
|
|
def load(self):
|
|
if self.model is None:
|
|
args = {}
|
|
if self.device == "cuda":
|
|
args.update(device_map="auto")
|
|
if self.load_8bit:
|
|
args.update(device_map="auto", load_in_8bit=True)
|
|
args.update(torch_dtype=getattr(torch, self.dtype) if self.dtype != "auto" else "auto")
|
|
self.model = AutoModelForCausalLM.from_pretrained(
|
|
self.model_path, trust_remote_code=self.trust_remote_code, **args
|
|
)
|
|
self.model.eval()
|
|
if "device_map" not in args:
|
|
self.model.to(self.device)
|
|
print_gpu_utilization()
|
|
# Sampling with temperature will cause MMLU to drop
|
|
self.model.generation_config.do_sample = False
|
|
if self.tokenizer is None:
|
|
self.tokenizer = AutoTokenizer.from_pretrained(
|
|
self.model_path, trust_remote_code=self.trust_remote_code
|
|
)
|
|
|
|
def run(self, prompt: str, **kwargs) -> str:
|
|
self.load()
|
|
assert self.model is not None, "Model must be loaded before running."
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded before running."
|
|
device = self.model.device if hasattr(self.model, "device") else self.device
|
|
inputs = self.tokenizer(prompt, return_tensors="pt").to(device)
|
|
if "RWForCausalLM" in str(type(self.model)) or "Falcon" in str(type(self.model)):
|
|
# this key is used by falcon 180b, but not by falcon 40b
|
|
inputs.pop("token_type_ids", None)
|
|
|
|
outputs = self.model.generate(
|
|
**inputs,
|
|
max_new_tokens=self.max_output_length,
|
|
pad_token_id=self.tokenizer.eos_token_id, # Avoid pad token warning
|
|
**kwargs,
|
|
)
|
|
batch_size, length = inputs.input_ids.shape
|
|
return self.tokenizer.decode(outputs[0, length:], skip_special_tokens=True)
|
|
|
|
def get_choice(self, text: str, **kwargs) -> tuple[float, float]:
|
|
self.load()
|
|
assert self.model is not None, "Model must be loaded to get choices."
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded to get choices."
|
|
inputs = self.tokenizer(text, return_tensors="pt").to(self.device)
|
|
with torch.no_grad():
|
|
predictions = self.model(
|
|
**inputs,
|
|
**kwargs,
|
|
).logits[0, -1]
|
|
a_index = self.tokenizer("A", add_special_tokens=False).input_ids[0]
|
|
b_index = self.tokenizer("B", add_special_tokens=False).input_ids[0]
|
|
a = float(predictions[a_index].cpu())
|
|
b = float(predictions[b_index].cpu())
|
|
return a, b
|
|
|
|
|
|
class AutoAWQCausalModel(SeqToSeqModel):
|
|
def load(self):
|
|
if self.model is None:
|
|
args = {}
|
|
if self.device == "cuda":
|
|
args.update(device_map="auto")
|
|
if self.load_8bit:
|
|
args.update(device_map="auto", load_in_8bit=True)
|
|
args.update(torch_dtype=getattr(torch, self.dtype) if self.dtype != "auto" else "auto")
|
|
self.model = AutoAWQForCausalLM.from_quantized(
|
|
self.model_path, trust_remote_code=self.trust_remote_code, **args
|
|
)
|
|
self.model.eval()
|
|
if "device_map" not in args:
|
|
self.model.to(self.device)
|
|
print_gpu_utilization()
|
|
# Sampling with temperature will cause MMLU to drop
|
|
self.model.config.do_sample = False
|
|
if self.tokenizer is None:
|
|
self.tokenizer = AutoTokenizer.from_pretrained(
|
|
self.model_path, trust_remote_code=self.trust_remote_code
|
|
)
|
|
|
|
def run(self, prompt: str, **kwargs) -> str:
|
|
self.load()
|
|
assert self.model is not None, "Model must be loaded before running."
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded before running."
|
|
device = self.model.device if hasattr(self.model, "device") else self.device
|
|
inputs = self.tokenizer(prompt, return_tensors="pt").to(device)
|
|
if "RWForCausalLM" in str(type(self.model)) or "Falcon" in str(type(self.model)):
|
|
# this key is used by falcon 180b, but not by falcon 40b
|
|
inputs.pop("token_type_ids", None)
|
|
|
|
outputs = self.model.generate(
|
|
**inputs,
|
|
max_new_tokens=self.max_output_length,
|
|
pad_token_id=self.tokenizer.eos_token_id, # Avoid pad token warning
|
|
**kwargs,
|
|
)
|
|
batch_size, length = inputs.input_ids.shape
|
|
return self.tokenizer.decode(outputs[0, length:], skip_special_tokens=True)
|
|
|
|
def get_choice(self, text: str, **kwargs) -> tuple[float, float]:
|
|
self.load()
|
|
assert self.model is not None, "Model must be loaded before running."
|
|
assert self.tokenizer is not None, "Tokenizer must be loaded before running."
|
|
inputs = self.tokenizer(text, return_tensors="pt").to(self.device)
|
|
with torch.no_grad():
|
|
predictions = self.model(
|
|
**inputs,
|
|
**kwargs,
|
|
).logits[0, -1]
|
|
a_index = self.tokenizer("A", add_special_tokens=False).input_ids[0]
|
|
b_index = self.tokenizer("B", add_special_tokens=False).input_ids[0]
|
|
a = float(predictions[a_index].cpu())
|
|
b = float(predictions[b_index].cpu())
|
|
return a, b
|
|
|
|
|
|
def print_gpu_utilization():
|
|
for i in range(torch.cuda.device_count()):
|
|
print(f"GPU {i}: {torch.cuda.memory_allocated(i) / 1e9} GB")
|
|
|
|
|
|
def select_model(model_name: str, **kwargs) -> EvalModel:
|
|
model_map = {
|
|
"causal": CausalModel,
|
|
"autoawq_causal": AutoAWQCausalModel,
|
|
}
|
|
model_class = model_map.get(model_name)
|
|
if model_class is None:
|
|
raise ValueError(f"{model_name}. Choose from {list(model_map.keys())}")
|
|
return model_class(**kwargs)
|
|
|
|
|
|
class TrtllmPipeline:
|
|
def __init__(self, tokenizer, model, model_name, pad_id, end_id, max_attention_window_size):
|
|
self.tokenizer = tokenizer
|
|
self.model = model
|
|
self.model_name = model_name
|
|
self.pad_id = pad_id
|
|
self.end_id = end_id
|
|
self.max_attention_window_size = max_attention_window_size
|
|
self.output_len = 2
|
|
|
|
def __call__(self, prompt):
|
|
rank = tensorrt_llm.mpi_rank()
|
|
# Run the model in batch size 1 and beam size 1
|
|
inputs = self.tokenizer.encode(prompt, return_tensors="pt").squeeze(0)
|
|
batch_input_ids: list[torch.Tensor] = [inputs]
|
|
|
|
# For multi-choice tasks like MMLU, we don't need to adjust following parameters
|
|
output_len = self.output_len
|
|
top_k = 1
|
|
top_p = 0.0
|
|
|
|
input_lengths = [x.size(0) for x in batch_input_ids]
|
|
|
|
with torch.no_grad():
|
|
if isinstance(self.model, nn.Module):
|
|
# Left padding for HF
|
|
max_length = max(input_lengths)
|
|
paddings = [
|
|
torch.ones(max_length - length, dtype=torch.int32) * self.pad_id
|
|
for length in input_lengths
|
|
]
|
|
batch_input_ids = [torch.cat([pad, x]) for x, pad in zip(batch_input_ids, paddings)]
|
|
batch_input_ids_tensor: torch.Tensor = torch.stack(batch_input_ids)
|
|
batch_input_ids_tensor = batch_input_ids_tensor.cuda()
|
|
with torch.no_grad():
|
|
# Use default temperature and top_k
|
|
outputs = self.model.generate(
|
|
batch_input_ids_tensor, max_new_tokens=output_len, top_k=top_k
|
|
)
|
|
output_ids = outputs[0, input_lengths[0] :]
|
|
|
|
elif isinstance(self.model, (ModelRunnerCpp, ModelRunner)):
|
|
outputs = self.model.generate(
|
|
batch_input_ids,
|
|
max_new_tokens=output_len,
|
|
max_attention_window_size=self.max_attention_window_size,
|
|
end_id=self.end_id,
|
|
pad_id=self.pad_id,
|
|
top_k=top_k,
|
|
top_p=top_p,
|
|
temperature=1,
|
|
)
|
|
torch.cuda.synchronize()
|
|
if rank == 0:
|
|
output_ids = outputs[0, 0, input_lengths[0] :]
|
|
|
|
if rank == 0:
|
|
return self.tokenizer.decode(output_ids, skip_special_tokens=True)
|
|
else:
|
|
return None
|
|
|
|
def check_valid_length(self, prompt):
|
|
if isinstance(self.model, nn.Module):
|
|
return True
|
|
input_len = len(self.tokenizer.encode(prompt))
|
|
return (
|
|
input_len <= self.model.max_input_len
|
|
and input_len + self.output_len <= self.model.max_seq_len
|
|
)
|
|
|
|
|
|
if __name__ == "__main__":
|
|
Fire()
|