Spaces:
Running
Running
File size: 11,384 Bytes
c374021 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 | """
experiments/cross_attention_patterns.py
========================================
Documents and compares the four distinct cross-attention (fusion) patterns
used by each architecture in this pipeline.
This module does NOT require loading any model β it produces a static
analysis table and inline architecture diagrams, and can optionally
compute the number of cross-attention parameter counts from loaded models.
Usage (standalone):
python -m experiments.cross_attention_patterns
Architecture Summary
--------------------
βββββββββββββββββββ¬ββββββββββββββββββββββββββββ¬βββββββββββββββββββββββββββββββββββ
β Architecture β Fusion Mechanism β Cross-Attention Exists? β
βββββββββββββββββββΌββββββββββββββββββββββββββββΌβββββββββββββββββββββββββββββββββββ€
β ViT-GPT2 β Standard Full CA β β
Yes β at every GPT-2 layer β
β BLIP (MED) β Gated Cross-Attention MED β β
Yes β between SA and FFN β
β GIT β Self-Attn Prefix β β No β unified causal SA β
β Custom VLM β Visual Prefix-Tuning β β No β linear projection + SA β
βββββββββββββββββββ΄ββββββββββββββββββββββββββββ΄βββββββββββββββββββββββββββββββββββ
"""
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
# Static Architecture Descriptions
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
PATTERNS = [
{
"name": "ViT-GPT2",
"model_id": "nlpconnect/vit-gpt2-image-captioning",
"cross_attention": True,
"ca_type": "Standard Full Cross-Attention",
"description": (
"Every GPT-2 decoder layer has an explicit cross-attention block. "
"Each text token attends to ALL 197 ViT patch embeddings "
"(1 CLS + 196 spatial) at every layer. "
"This is the brute-force approach β maximum information, highest compute."
),
"fusion_formula": "h_text = CrossAttn(Q=h_text, K=h_vis, V=h_vis)",
"ablation_support": True,
"ablation_method": "encoder_attention_mask on generate()",
},
{
"name": "BLIP (MED)",
"model_id": "Salesforce/blip-image-captioning-base",
"cross_attention": True,
"ca_type": "Gated Multimodal Encoder-Decoder (MED)",
"description": (
"BLIP's MED architecture injects a cross-attention sub-layer "
"BETWEEN the self-attention and FFN sub-layers at each decoder block. "
"A learnable gate controls how much visual information passes through. "
"This is more targeted than ViT-GPT2's brute-force attention."
),
"fusion_formula": (
"h = SA(h_text) "
"β h = h + gate * CrossAttn(Q=h, K=h_vis, V=h_vis) "
"β h = FFN(h)"
),
"ablation_support": True,
"ablation_method": "encoder_attention_mask via generate_with_mask()",
},
{
"name": "GIT",
"model_id": "microsoft/git-base-coco",
"cross_attention": False,
"ca_type": "Zero Cross-Attention (Self-Attention Prefix)",
"description": (
"GIT concatenates image patch embeddings directly in front of text tokens "
"to form a flat joint sequence: [img_tokens | text_tokens]. "
"A single causal self-attention Transformer processes the whole thing. "
"There is NO dedicated cross-attention block. "
"Modality fusion is implicit via positional self-attention."
),
"fusion_formula": "h = CausalSA([h_vis; h_text])",
"ablation_support": False,
"ablation_method": "N/A β no encoder_attention_mask concept",
},
{
"name": "Custom VLM (Shakespeare)",
"model_id": "google/vit-base-patch16-224-in21k (ViT) + char-level decoder",
"cross_attention": False,
"ca_type": "Visual Prefix-Tuning (Linear Bridge + Causal SA)",
"description": (
"A frozen ViT extracts 197 patch embeddings (768-dim). "
"A single trainable Linear(768β384) projects these to the decoder's "
"embedding space. Projected visual tokens are prepended to character "
"embeddings and the Shakespeare causal decoder processes them jointly. "
"Only the linear projection is trained (~294K params, <0.2% of total). "
"\nKey insight: cross-attention is provably unnecessary when modalities "
"are aligned in the same embedding space via prefix concatenation."
),
"fusion_formula": (
"v = Linear(ViT(img)) "
"β x = CausalSA([v; char_emb]) "
"β logits = LMHead(x[len(v):])"
),
"ablation_support": False,
"ablation_method": "N/A β visual prefix is part of unified sequence",
},
]
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
# Comparison Table Printer
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
def print_comparison_table():
"""Print a formatted comparison table to stdout."""
print("\n" + "=" * 80)
print(" Cross-Attention Pattern Comparison")
print("=" * 80)
print(f" {'Architecture':<22} {'CA?':>5} {'Type':<35} {'Ablation?':>9}")
print(" " + "-" * 76)
for p in PATTERNS:
ca = " β
" if p["cross_attention"] else " β"
abl = " β
" if p["ablation_support"] else " β"
print(f" {p['name']:<22} {ca:>5} {p['ca_type']:<35} {abl:>9}")
print("=" * 80)
for p in PATTERNS:
print(f"\n ββ {p['name']} ββββββββββββββββββββββββββββββββββββββββββββββ")
print(f" Model : {p['model_id']}")
print(f" CA Type: {p['ca_type']}")
print(f" Formula: {p['fusion_formula']}")
for line in p["description"].split("\n"):
print(f" {line.strip()}")
if p["ablation_support"]:
print(f" Ablation: {p['ablation_method']}")
else:
print(f" β οΈ Ablation: {p['ablation_method']}")
print()
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
# Optional: Parameter Count from Loaded Models
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
def count_cross_attention_params(model, model_name: str) -> dict:
"""
Count parameters in cross-attention layers for BLIP or ViT-GPT2.
For GIT / Custom VLM (no CA), returns zero.
Args:
model : loaded PyTorch model
model_name : 'blip' | 'vit_gpt2' | 'git' | 'custom'
Returns:
dict with 'total', 'cross_attn', 'cross_attn_pct'
"""
total = sum(p.numel() for p in model.parameters())
ca_params = 0
if model_name == "blip":
for name, p in model.named_parameters():
if "crossattention" in name.lower():
ca_params += p.numel()
elif model_name == "vit_gpt2":
for name, p in model.named_parameters():
if "crossattention" in name.lower() or "cross_attn" in name.lower():
ca_params += p.numel()
# GIT / custom: 0 cross-attention params by design
return {
"model": model_name,
"total_params": total,
"cross_attn_params": ca_params,
"cross_attn_pct": ca_params / total * 100 if total > 0 else 0.0,
}
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
# CLI
# βββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββββ
def main():
print_comparison_table()
# Optionally count params for all four models
count_params = input(
"\nCount cross-attention parameters in all models? "
"(requires downloading BLIP+ViT-GPT2+GIT) [y/N]: "
).strip().lower()
if count_params == "y":
import torch
device = torch.device("cpu")
print("\nLoading models to count parameters...\n")
import sys, os
sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
from config import CFG
from models.blip_tuner import get_blip_model
from models.vit_gpt2_tuner import get_vit_gpt2_model
from models.git_tuner import get_git_model
from models.custom_vlm import CustomVLM, build_char_vocab
cfg = CFG()
rows = []
model_b, _ = get_blip_model(cfg, device)
rows.append(count_cross_attention_params(model_b, "blip"))
del model_b
model_v, _, _ = get_vit_gpt2_model(cfg, device)
rows.append(count_cross_attention_params(model_v, "vit_gpt2"))
del model_v
model_g, _ = get_git_model(cfg, device)
rows.append(count_cross_attention_params(model_g, "git"))
del model_g
with open(cfg.shakespeare_file, "r") as f:
text = f.read()
_, c2i, i2c, vs = build_char_vocab(text)
model_c = CustomVLM(vocab_size=vs)
rows.append(count_cross_attention_params(model_c, "custom"))
del model_c
print("\n" + "=" * 65)
print(" Cross-Attention Parameter Counts")
print("=" * 65)
print(f" {'Model':<15} {'Total':>12} {'CA Params':>12} {'CA %':>8}")
print(" " + "-" * 58)
for r in rows:
print(f" {r['model']:<15} {r['total_params']:>12,} "
f"{r['cross_attn_params']:>12,} {r['cross_attn_pct']:>7.2f}%")
print("=" * 65)
if __name__ == "__main__":
main()
|