Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
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
1 change: 1 addition & 0 deletions pyproject.toml
Original file line number Diff line number Diff line change
Expand Up @@ -58,6 +58,7 @@ dependencies = [
"pydantic",
"supervision",
"matplotlib",
"soft_moe",
]

[project.optional-dependencies]
Expand Down
2 changes: 2 additions & 0 deletions rfdetr/config.py
Original file line number Diff line number Diff line change
Expand Up @@ -30,6 +30,8 @@ class ModelConfig(BaseModel):
resolution: int = 560
group_detr: int = 13
gradient_checkpointing: bool = False
MoE: bool = False
MoE_params: List[int] = [32, 1]

class RFDETRBaseConfig(ModelConfig):
encoder: Literal["dinov2_windowed_small", "dinov2_windowed_base"] = "dinov2_windowed_small"
Expand Down
63 changes: 53 additions & 10 deletions rfdetr/models/transformer.py
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,7 @@
from typing import Optional

import torch
from soft_moe import SoftMoELayerWrapper
import torch.nn.functional as F
from torch import nn, Tensor

Expand All @@ -39,6 +40,18 @@ def forward(self, x):
x = F.relu(layer(x)) if i < self.num_layers - 1 else layer(x)
return x

class FFNBlock(nn.Module):
def __init__(self, d_model, dim_feedforward, dropout):
super().__init__()
self.net = nn.Sequential(
nn.Linear(d_model, dim_feedforward),
nn.ReLU(),
nn.Dropout(dropout),
nn.Linear(dim_feedforward, d_model),
)

def forward(self, x):
return self.net(x)

def gen_sineembed_for_position(pos_tensor, dim=128):
# n_query, bs, _ = pos_tensor.size()
Expand Down Expand Up @@ -136,7 +149,8 @@ def __init__(self, d_model=512, sa_nhead=8, ca_nhead=8, num_queries=300,
num_feature_levels=4, dec_n_points=4,
lite_refpoint_refine=False,
decoder_norm_type='LN',
bbox_reparam=False):
bbox_reparam=False,
MoE=False, MoE_params=[32,1]):
super().__init__()
self.encoder = None

Expand All @@ -145,7 +159,9 @@ def __init__(self, d_model=512, sa_nhead=8, ca_nhead=8, num_queries=300,
group_detr=group_detr,
num_feature_levels=num_feature_levels,
dec_n_points=dec_n_points,
skip_self_attn=False,)
skip_self_attn=False,
MoE=MoE,
MoE_params=MoE_params)
assert decoder_norm_type in ['LN', 'Identity']
norm = {
"LN": lambda channels: nn.LayerNorm(channels),
Expand Down Expand Up @@ -441,7 +457,7 @@ class TransformerDecoderLayer(nn.Module):
def __init__(self, d_model, sa_nhead, ca_nhead, dim_feedforward=2048, dropout=0.1,
activation="relu", normalize_before=False, group_detr=1,
num_feature_levels=4, dec_n_points=4,
skip_self_attn=False):
skip_self_attn=False, MoE=False, MoE_params=[32,1]):
super().__init__()
# Decoder Self-Attention
self.self_attn = nn.MultiheadAttention(embed_dim=d_model, num_heads=sa_nhead, dropout=dropout, batch_first=True)
Expand All @@ -453,19 +469,41 @@ def __init__(self, d_model, sa_nhead, ca_nhead, dim_feedforward=2048, dropout=0.
d_model, n_levels=num_feature_levels, n_heads=ca_nhead, n_points=dec_n_points)

self.nhead = ca_nhead

# Implementation of Feedforward model
self.linear1 = nn.Linear(d_model, dim_feedforward)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(dim_feedforward, d_model)

# Implementation of Feedforward or the MoE Layer (done by @LeosCtrt)
self.MoE = MoE
if self.MoE == True:
print("\n" + "="*80)
print("Loading Mixture of Expert (MoE) Architecture")
print("="*80)
print(f"Experts Count : {MoE_params[0]}")
print(f"Slots per Expert : {MoE_params[1]}")
print("-"*80)
print("Warning: This custom architecture prevents loading full pretrained weights.")
print("Note : It may be slightly slower but could improve accuracy.")
print("="*80 + "\n")

self.moe_layer = SoftMoELayerWrapper(
dim=d_model,
num_experts=MoE_params[0],
slots_per_expert=MoE_params[1],
layer=FFNBlock,
d_model=d_model,
dim_feedforward=dim_feedforward,
dropout=dropout
)
else:
self.linear1 = nn.Linear(d_model, dim_feedforward)
self.dropout = nn.Dropout(dropout)
self.linear2 = nn.Linear(dim_feedforward, d_model)
self.activation = _get_activation_fn(activation)

self.norm2 = nn.LayerNorm(d_model)
self.norm3 = nn.LayerNorm(d_model)

self.dropout2 = nn.Dropout(dropout)
self.dropout3 = nn.Dropout(dropout)

self.activation = _get_activation_fn(activation)
self.normalize_before = normalize_before
self.group_detr = group_detr

Expand Down Expand Up @@ -521,7 +559,10 @@ def forward_post(self, tgt, memory,

tgt = tgt + self.dropout2(tgt2)
tgt = self.norm2(tgt)
tgt2 = self.linear2(self.dropout(self.activation(self.linear1(tgt))))
if self.MoE == True:
tgt2 = self.moe_layer(tgt)
else:
tgt2 = self.linear2(self.dropout(self.activation(self.linear1(tgt))))
tgt = tgt + self.dropout3(tgt2)
tgt = self.norm3(tgt)
return tgt
Expand Down Expand Up @@ -571,6 +612,8 @@ def build_transformer(args):
lite_refpoint_refine=args.lite_refpoint_refine,
decoder_norm_type=args.decoder_norm,
bbox_reparam=args.bbox_reparam,
MoE=args.MoE,
MoE_params=args.MoE_params
)


Expand Down