model-pruning

モデルプルーニング：LLM の圧縮

このスキルを使うタイミング

モデルプルーニングは、次のような場合に使用します:

• 精度低下 1% 未満でモデルサイズを 40〜60% 削減 したいとき
• ハードウェアに適したスパース性を利用して推論を高速化（2〜4× の高速化）したいとき
• 制約のあるハードウェア（モバイル、エッジデバイス）にデプロイしたいとき
• ワンショット手法を使い、再学習なしで圧縮したいとき
• メモリフットプリントを削減して効率的なサービングを実現したいとき

主要な手法: Wanda（重み × アクティベーション）、SparseGPT（二次情報を利用）、構造化プルーニング、N:M スパース性

論文: Wanda ICLR 2024（arXiv 2306.11695）、SparseGPT（arXiv 2301.00774）

インストール

# Wanda implementation
git clone https://github.com/locuslab/wanda
cd wanda
pip install -r requirements.txt

# Optional: SparseGPT
git clone https://github.com/IST-DASLab/sparsegpt
cd sparsegpt
pip install -e .

# Dependencies
pip install torch transformers accelerate

クイックスタート

Wanda プルーニング（ワンショット、再学習なし）

出典: ICLR 2024（arXiv 2306.11695）

import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

# Load model
model = AutoModelForCausalLM.from_pretrained(
    "meta-llama/Llama-2-7b-hf",
    torch_dtype=torch.float16,
    device_map="cuda"
)
tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-7b-hf")

# Calibration data (small dataset for activation statistics)
calib_data = [
    "The quick brown fox jumps over the lazy dog.",
    "Machine learning is transforming the world.",
    "Artificial intelligence powers modern applications.",
]

# Wanda pruning function
def wanda_prune(model, calib_data, sparsity=0.5):
    """
    Wanda: Prune by weight magnitude × input activation.

    Args:
        sparsity: Fraction of weights to prune (0.5 = 50%)
    """
    # 1. Collect activation statistics
    activations = {}

    def hook_fn(name):
        def hook(module, input, output):
            # Store input activation norms
            activations[name] = input[0].detach().abs().mean(dim=0)
        return hook

    # Register hooks for all linear layers
    hooks = []
    for name, module in model.named_modules():
        if isinstance(module, torch.nn.Linear):
            hooks.append(module.register_forward_hook(hook_fn(name)))

    # Run calibration data
    model.eval()
    with torch.no_grad():
        for text in calib_data:
            inputs = tokenizer(text, return_tensors="pt").to(model.device)
            model(**inputs)

    # Remove hooks
    for hook in hooks:
        hook.remove()

    # 2. Prune weights based on |weight| × activation
    for name, module in model.named_modules():
        if isinstance(module, torch.nn.Linear) and name in activations:
            W = module.weight.data
            act = activations[name]

            # Compute importance: |weight| × activation
            importance = W.abs() * act.unsqueeze(0)

            # Flatten and find threshold
            threshold = torch.quantile(importance.flatten(), sparsity)

            # Create mask
            mask = importance >= threshold

            # Apply mask (prune)
            W *= mask.float()

    return model

# Apply Wanda pruning (50% sparsity, one-shot, no retraining)
pruned_model = wanda_prune(model, calib_data, sparsity=0.5)

# Save
pruned_model.save_pretrained("./llama-2-7b-wanda-50")

SparseGPT（二次情報を利用したプルーニング）

出典: arXiv 2301.00774

from sparsegpt import SparseGPT

# Load model
model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-2-7b-hf")

# Initialize SparseGPT
pruner = SparseGPT(model)

# Calibration data
calib_data = load_calibration_data()  # ~128 samples

# Prune (one-shot, layer-wise reconstruction)
pruned_model = pruner.prune(
    calib_data=calib_data,
    sparsity=0.5,           # 50% sparsity
    prunen=0,               # Unstructured (0) or N:M structured
    prunem=0,
    percdamp=0.01,          # Damping for Hessian inverse
)

# Results: Near-lossless pruning at 50% sparsity

N:M 構造化プルーニング（ハードウェアアクセラレータ向け）

def nm_prune(weight, n=2, m=4):
    """
    N:M pruning: Keep N weights per M consecutive weights.
    Example: 2:4 = keep 2 out of every 4 weights.

    Compatible with NVIDIA sparse tensor cores (2:4, 4:8).
    """
    # Reshape weight into groups of M
    shape = weight.shape
    weight_flat = weight.flatten()

    # Pad to multiple of M
    pad_size = (m - weight_flat.numel() % m) % m
    weight_padded = F.pad(weight_flat, (0, pad_size))

    # Reshape into (num_groups, m)
    weight_grouped = weight_padded.reshape(-1, m)

    # Find top-N in each group
    _, indices = torch.topk(weight_grouped.abs(), n, dim=-1)

    # Create mask
    mask = torch.zeros_like(weight_grouped)
    mask.scatter_(1, indices, 1.0)

    # Apply mask
    weight_pruned = weight_grouped * mask

    # Reshape back
    weight_pruned = weight_pruned.flatten()[:weight_flat.numel()]
    return weight_pruned.reshape(shape)

# Apply 2:4 sparsity (NVIDIA hardware)
for name, module in model.named_modules():
    if isinstance(module, torch.nn.Linear):
        module.weight.data = nm_prune(module.weight.data, n=2, m=4)

# 50% sparsity, 2× speedup on A100 with sparse tensor cores

コアコンセプト

1. プルーニング基準

マグニチュードプルーニング（ベースライン）:

# Prune weights with smallest absolute values
importance = weight.abs()
threshold = torch.quantile(importance, sparsity)
mask = importance >= threshold

Wanda（重み × アクティベーション）:

# Importance = |weight| × input_activation
importance = weight.abs() * activation
# Better than magnitude alone (considers usage)

SparseGPT（二次情報を利用）:

# Uses Hessian (second derivative) for importance
# More accurate but computationally expensive
importance = weight^2 / diag(Hessian)

2. 構造化 vs 非構造化

非構造化（細粒度）:

• 個々の重みをプルーニングする
• 品質が高い（精度が良い）
• ハードウェアによる高速化はなし（不規則なスパース性のため）

構造化（粗粒度）:

• ニューロン、ヘッド、またはレイヤー全体をプルーニングする
• 品質が低い（精度低下が大きい）
• ハードウェアによる高速化あり（規則的なスパース性のため）

準構造化（N:M）:

• 両者の長所を兼ね備える
• 50% スパース性（2:4）→ NVIDIA GPU で 2× の高速化
• 精度低下は最小限

3. スパース性のパターン

# Unstructured (random)
# [1, 0, 1, 0, 1, 1, 0, 0]
# Pros: Flexible, high quality
# Cons: No speedup

# Structured (block)
# [1, 1, 0, 0, 1, 1, 0, 0]
# Pros: Hardware friendly
# Cons: More accuracy loss

# N:M (semi-structured)
# [1, 0, 1, 0] [1, 1, 0, 0]  (2:4 pattern)
# Pros: Hardware speedup + good quality
# Cons: Requires specific hardware (NVIDIA)

プルーニング戦略

戦略 1: 段階的マグニチュードプルーニング

def gradual_prune(model, initial_sparsity=0.0, final_sparsity=0.5, num_steps=100):
    """Gradually increase sparsity during training."""
    for step in range(num_steps):
        # Current sparsity
        current_sparsity = initial_sparsity + (final_sparsity - initial_sparsity) * (step / num_steps)

        # Prune at current sparsity
        for module in model.modules():
            if isinstance(module, torch.nn.Linear):
                weight = module.weight.data
                threshold = torch.quantile(weight.abs().flatten(), current_sparsity)
                mask = weight.abs() >= threshold
                weight *= mask.float()

        # Train one step
        train_step(model)

    return model

戦略 2: レイヤーごとのプルーニング

def layer_wise_prune(model, sparsity_per_layer):
    """Different sparsity for different layers."""
    # Early layers: Less pruning (more important)
    # Late layers: More pruning (less critical)

    sparsity_schedule = {
        "layer.0": 0.3,   # 30% sparsity
        "layer.1": 0.4,
        "layer.2": 0.5,
        "layer.3": 0.6,   # 60% sparsity
    }

    for name, module in model.named_modules():
        if isinstance(module, torch.nn.Linear):
            # Find layer index
            for layer_name, sparsity in sparsity_schedule.items():
                if layer_name in name:
                    # Prune at layer-specific sparsity
                    prune_layer(module, sparsity)
                    break

    return model

戦略 3: 反復プルーニング + ファインチューニング

def iterative_prune_finetune(model, target_sparsity=0.5, iterations=5):
    """Prune gradually with fine-tuning between iterations."""
    current_sparsity = 0.0
    sparsity_increment = target_sparsity / iterations

    for i in range(iterations):
        # Increase sparsity
        current_sparsity += sparsity_increment

        # Prune
        prune_model(model, sparsity=current_sparsity)

        # Fine-tune (recover accuracy)
        fine_tune(model, epochs=2, lr=1e-5)

    return model

# Results: Better accuracy than one-shot at high sparsity

本番環境へのデプロイ

完全なプルーニングパイプライン

from transformers import Trainer, TrainingArguments

def production_pruning_pipeline(
    model_name="meta-llama/Llama-2-7b-hf",
    target_sparsity=0.5,
    method="wanda",  # or "sparsegpt"
):
    # 1. Load model
    model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16)
    tokenizer = AutoTokenizer.from_pretrained(model_name)

    # 2. Load calibration data
    calib_dataset = load_dataset("wikitext", "wikitext-2-raw-v1", split="train[:1000]")

    # 3. Apply pruning
    if method == "wanda":
        pruned_model = wanda_prune(model, calib_dataset, sparsity=target_sparsity)
    elif method == "sparsegpt":
        pruner = SparseGPT(model)
        pruned_model = pruner.prune(calib_dataset, sparsity=target_sparsity)

    # 4. (Optional) Fine-tune to recover accuracy
    training_args = TrainingArguments(
        output_dir="./pruned-model",
        num_train_epochs=1,
        per_device_train_batch_size=4,
        learning_rate=1e-5,
        bf16=True,
    )

    trainer = Trainer(
        model=pruned_model,
        args=training_args,
        train_dataset=finetune_dataset,
    )

    trainer.train()

    # 5. Save
    pruned_model.save_pretrained("./pruned-llama-7b-50")
    tokenizer.save_pretrained("./pruned-llama-7b-50")

    return pruned_model

# Usage
pruned_model = production_pruning_pipeline(
    model_name="meta-llama/Llama-2-7b-hf",
    target_sparsity=0.5,
    method="wanda"
)

評価

from lm_eval import evaluator

# Evaluate pruned vs original model
original_results = evaluator.simple_evaluate(
    model="hf",
    model_args="pretrained=meta-llama/Llama-2-7b-hf",
    tasks=["arc_easy", "hellaswag", "winogrande"],
)

pruned_results = evaluator.simple_evaluate(
    model="hf",
    model_args="pretrained=./pruned-llama-7b-50",
    tasks=["arc_easy", "hellaswag", "winogrande"],
)

# Compare
print(f"Original: {original_results['results']['arc_easy']['acc']:.3f}")
print(f"Pruned:   {pruned_results['results']['arc_easy']['acc']:.3f}")
print(f"Degradation: {(original_results - pruned_results):.3f}")

# Typical results at 50% sparsity:
# - Wanda: <1% accuracy loss
# - SparseGPT: <0.5% accuracy loss
# - Magnitude: 2-3% accuracy loss

ベストプラクティス

1. スパース性の選択

# Conservative (safe)
sparsity = 0.3  # 30%, <0.5% loss

# Balanced (recommended)
sparsity = 0.5  # 50%, ~1% loss

# Aggressive (risky)
sparsity = 0.7  # 70%, 2-5% loss

# Extreme (model-dependent)
sparsity = 0.9  # 90%, significant degradation

2. 手法の選択

# One-shot, no retraining → Wanda or SparseGPT
if no_retraining_budget:
    use_method = "wanda"  # Faster

# Best quality → SparseGPT
if need_best_quality:
    use_method = "sparsegpt"  # More accurate

# Hardware speedup → N:M structured
if need_speedup:
    use_method = "nm_prune"  # 2:4 or 4:8

3. よくある落とし穴の回避

# ❌ Bad: Pruning without calibration data
prune_random(model)  # No activation statistics

# ✅ Good: Use calibration data
prune_wanda(model, calib_data)

# ❌ Bad: Too high sparsity in one shot
prune(model, sparsity=0.9)  # Massive accuracy loss

# ✅ Good: Gradual or iterative
iterative_prune(model, target=0.9, steps=10)

性能比較

50% スパース性でのプルーニング手法（LLaMA-7B）:

手法	精度低下	速度	メモリ	再学習の必要性
Magnitude	-2.5%	1.0×	-50%	なし
Wanda	-0.8%	1.0×	-50%	なし
SparseGPT	-0.4%	1.0×	-50%	なし
N:M (2:4)	-1.0%	2.0×	-50%	なし
Structured	-3.0%	2.0×	-50%	なし

出典: Wanda 論文（ICLR 2024）、SparseGPT 論文

リソース

• Wanda 論文（ICLR 2024）: https://arxiv.org/abs/2306.11695
• Wanda GitHub: https://github.com/locuslab/wanda
• SparseGPT 論文: https://arxiv.org/abs/2301.00774
• SparseGPT GitHub: https://github.com/IST-DASLab/sparsegpt
• NVIDIA Sparse Tensor Cores: https://developer.nvidia.com/blog/accelerating-inference-with-sparsity-using-ampere-and-tensorrt/