Fix unfused QKV params case; stack vs interleave option#83

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
ksivaman merged 2 commits into from
Mar 2, 2023
Merged

Fix unfused QKV params case; stack vs interleave option #83

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
b5d4ba6
fix qkv weight unfused path
ksivaman Mar 1, 2023
f561579
fix non FA non interleaved case
ksivaman Mar 1, 2023
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
79 changes: 59 additions & 20 deletions transformer_engine/pytorch/transformer.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
divide,
attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
cast_if_needed,
get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
)

# [sq, b, np, hn] -> [sq, b * np, hn]
query_layer = query_layer.view(
query_layer = query_layer.reshape(
output_size[2], output_size[0] * output_size[1], -1
)
# [sk, b, np, hn] -> [sk, b * np, hn]
key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

# preallocting result tensor: [b * np, sq, sk]
matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
)

# change view [sk, b * np, hn]
value_layer = value_layer.view(
value_layer = value_layer.reshape(
value_layer.size(0), output_size[0] * output_size[1], -1
)

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
set_parallel_mode: bool = False,
fuse_qkv_params: bool = False,
zero_centered_gamma: bool = False,
qkv_weight_interleaved: bool = True,
) -> None:
super().__init__()
self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
self.params_dtype = params_dtype
self.init_method = init_method

if not fuse_qkv_params:
qkv_weight_interleaved = False
self.qkv_weight_interleaved = qkv_weight_interleaved

assert (
attention_type in AttnTypes
), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
is_first_microbatch=is_first_microbatch,
)

# [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
new_tensor_shape = mixed_x_layer.size()[:-1] + (
self.num_attention_heads_per_partition,
3 * self.hidden_size_per_attention_head,
)
if self.qkv_weight_interleaved:
# [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
new_tensor_shape = mixed_x_layer.size()[:-1] + (
self.num_attention_heads_per_partition,
3 * self.hidden_size_per_attention_head,
)
# split along last dimension
split_dim = -1
else:
# [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
new_tensor_shape = mixed_x_layer.size()[:-1] + (
3 * self.num_attention_heads_per_partition,
self.hidden_size_per_attention_head,
)
# split along second last dimension
split_dim = -2

mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

# [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
query_layer, key_layer, value_layer = split_tensor_along_last_dim(
mixed_x_layer, 3
# mixed_x_layer --> 3 [sq, b, np, hn]
query_layer, key_layer, value_layer = split_tensor_along_dim(
mixed_x_layer, split_dim, 3
)
else:
# Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
is_first_microbatch=is_first_microbatch,
)

# [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
new_tensor_shape = mixed_kv_layer.size()[:-1] + (
self.num_attention_heads_per_partition,
2 * self.hidden_size_per_attention_head,
)
if self.qkv_weight_interleaved:
# [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
new_tensor_shape = mixed_kv_layer.size()[:-1] + (
self.num_attention_heads_per_partition,
2 * self.hidden_size_per_attention_head,
)
# split along last dimension
split_dim = -1
else:
# [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
new_tensor_shape = mixed_kv_layer.size()[:-1] + (
2 * self.num_attention_heads_per_partition,
self.hidden_size_per_attention_head,
)
# split along second last dimension
split_dim = -2

mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

# [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
# mixed_kv_layer --> 2 [sk, b, np, hn]
key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

# Attention head [sq, b, h] --> [sq, b, hp]
if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
.. math::
y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
(1 + \gamma) + \beta

qkv_weight_interleaved : bool, default = `True`
if set to `False`, the QKV weight is interpreted as a concatenation of
query, key, and value weights along the `0th` dimension. The default
interpretation is that the individual `q`, `k`, and `v` weights for each
attention head are interleaved. This parameter is set to `False` when
using :attr:`fuse_qkv_params=False`.
Parallelism parameters
----------------------
set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
set_parallel_mode: bool = False,
fuse_qkv_params: bool = False,
zero_centered_gamma: bool = False,
qkv_weight_interleaved: bool = True,
) -> None:
super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
not fuse_wgrad_accumulation
), "Gradient accumulation fusion requires single QKV parameter."

if not fuse_qkv_params:
qkv_weight_interleaved = False

self.kv_channels = (
kv_channels if kv_channels else (hidden_size // num_attention_heads)
)
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
"set_parallel_mode": set_parallel_mode,
"fuse_qkv_params": fuse_qkv_params,
"zero_centered_gamma": zero_centered_gamma,
"qkv_weight_interleaved" : qkv_weight_interleaved,
}

self.self_attention = MultiHeadAttention(
Expand Down
9 changes: 4 additions & 5 deletions transformer_engine/pytorch/utils.py
View file
Open in desktop
Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
return numerator // denominator


def split_tensor_along_last_dim(
tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
"""Split a tensor along its last dimension.
Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
in memory.
"""
# Get the size and dimension.
last_dim = tensor.dim() - 1
last_dim_size = divide(tensor.size()[last_dim], num_partitions)
split_size = divide(tensor.size()[dim], num_partitions)
# Split.
tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
tensor_list = torch.split(tensor, split_size, dim=dim)
# Note: torch.split does not create contiguous tensors by default.
if contiguous_split_chunks:
return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down
, 'i'); if (__m === '*' || __re.test(location.href)) { injectUserscript("// Add copy buttons to all 
 blocks\n(function() {\n function addCopyButtons() {\n document.querySelectorAll('pre code').forEach(function(codeBlock) {\n if (codeBlock.parentElement.hasAttribute('data-copy-added')) return;\n codeBlock.parentElement.setAttribute('data-copy-added', 'true');\n \n var btn = document.createElement('button');\n btn.textContent = 'Copy';\n btn.style.cssText = 'position:absolute;top:4px;right:4px;padding:2px 8px;font-size:11px;background:#4ecdc4;border:none;border-radius:4px;color:#1a1a2e;cursor:pointer;opacity:0.7;transition:opacity 0.2s;';\n btn.onmouseover = function() { this.style.opacity = '1'; };\n btn.onmouseout = function() { this.style.opacity = '0.7'; };\n btn.onclick = function() {\n navigator.clipboard.writeText(codeBlock.textContent).then(function() {\n btn.textContent = 'Copied!';\n setTimeout(function() { btn.textContent = 'Copy'; }, 1500);\n });\n };\n codeBlock.parentElement.style.position = 'relative';\n codeBlock.parentElement.appendChild(btn);\n });\n }\n \n addCopyButtons();\n \n // Re-run on dynamic content\n var observer = new MutationObserver(addCopyButtons);\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "Add Copy Buttons to Code Blocks");
}
} catch(__e) { console.warn('[Userscript:Add Copy Buttons to Code Blocks]', __e); }
})();
(function(){
try {
var __m = "github.com";
var __re = new RegExp('^' + "github\\.com" + '

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Force GitHub README to respect dark mode\n(function() {\n var style = document.createElement('style');\n style.textContent = '\n .markdown-body {\n color-scheme: dark light;\n }\n .markdown-body pre { background: #161b22 !important; }\n .markdown-body code { background: rgba(110, 118, 129, 0.4) !important; }\n .markdown-body table th, .markdown-body table td { border-color: #30363d !important; }\n .markdown-body img { background: #0d1117; }\n .markdown-body blockquote { border-left-color: #8b949e; }\n .markdown-body hr { border-color: #30363d; }\n ';\n document.head.appendChild(style);\n})();", "GitHub Dark Mode README Fix");
}
} catch(__e) { console.warn('[Userscript:GitHub Dark Mode README Fix]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Highlight search terms from Google/DuckDuckGo/Bing referrer\n(function() {\n var ref = document.referrer;\n var terms = [];\n \n if (ref.includes('google.com') || ref.includes('duckduckgo.com') || ref.includes('bing.com')) {\n var url = new URL(ref);\n var q = url.searchParams.get('q') || url.searchParams.get('p');\n if (q) {\n terms = q.split(/\\s+/).filter(function(t) { return t.length > 2; });\n }\n }\n \n if (terms.length === 0) return;\n \n var style = document.createElement('style');\n style.textContent = '.userscript-highlight { background: #fbbf24; color: #1a1a2e; padding: 1px 3px; border-radius: 2px; }';\n document.head.appendChild(style);\n \n function highlight(node) {\n if (node.nodeType === 3) { // text node\n var text = node.textContent;\n var found = false;\n terms.forEach(function(term) {\n var regex = new RegExp('(' + term.replace(/[.*+?^${}()|[\\]\\\\]/g, '\\\\') + ')', 'gi');\n if (regex.test(text)) {\n found = true;\n var frag = document.createDocumentFragment();\n var parts = text.split(regex);\n parts.forEach(function(part, i) {\n if (i % 2 === 0) {\n frag.appendChild(document.createTextNode(part));\n } else {\n var span = document.createElement('span');\n span.className = 'userscript-highlight';\n span.textContent = part;\n frag.appendChild(span);\n }\n });\n node.parentNode.replaceChild(frag, node);\n }\n });\n } else if (node.nodeType === 1 && node.childNodes) { // element\n var skipTags = ['SCRIPT', 'STYLE', 'NOSCRIPT', 'TEXTAREA', 'INPUT', 'SELECT'];\n if (!skipTags.includes(node.tagName)) {\n Array.from(node.childNodes).forEach(highlight);\n }\n }\n }\n \n highlight(document.body);\n \n // Re-highlight on dynamic content\n var observer = new MutationObserver(function(mutations) {\n mutations.forEach(function(m) {\n m.addedNodes.forEach(function(node) {\n if (node.nodeType === 1 || node.nodeType === 3) highlight(node);\n });\n });\n });\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "Highlight Search Terms");
}
} catch(__e) { console.warn('[Userscript:Highlight Search Terms]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Strip utm_, fbclid, gclid, etc. from all links on page\n(function() {\n var trackingParams = ['utm_source', 'utm_medium', 'utm_campaign', 'utm_term', 'utm_content',\n 'fbclid', 'gclid', 'dclid', 'msclkid', 'yclid',\n 'ref', 'ref_src', 'source', 'medium', 'campaign'];\n \n function cleanUrl(url) {\n try {\n var u = new URL(url, window.location.origin);\n var changed = false;\n trackingParams.forEach(function(p) {\n if (u.searchParams.has(p)) {\n u.searchParams.delete(p);\n changed = true;\n }\n });\n return changed ? u.toString() : url;\n } catch (e) {\n return url;\n }\n }\n \n function cleanLinks() {\n document.querySelectorAll('a[href]').forEach(function(a) {\n var clean = cleanUrl(a.href);\n if (clean !== a.href) a.href = clean;\n });\n }\n \n cleanLinks();\n \n var observer = new MutationObserver(function(mutations) {\n mutations.forEach(function(m) {\n m.addedNodes.forEach(function(node) {\n if (node.nodeType === 1) {\n if (node.tagName === 'A') cleanLinks();\n node.querySelectorAll('a[href]').forEach(function(a) {\n var clean = cleanUrl(a.href);\n if (clean !== a.href) a.href = clean;\n });\n }\n });\n });\n });\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "Remove Tracking Parameters from Links");
}
} catch(__e) { console.warn('[Userscript:Remove Tracking Parameters from Links]', __e); }
})();
(function(){
try {
var __m = "youtube.com";
var __re = new RegExp('^' + "youtube\\.com" + '

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Auto-enable theater mode on YouTube\n(function() {\n function tryTheater() {\n var btn = document.querySelector('button[aria-label=\"Theater mode\"], ytd-player #player button[title=\"Theater mode\"]');\n if (btn && !btn.classList.contains('activated')) {\n btn.click();\n }\n }\n \n // Try immediately\n tryTheater();\n \n // Try after navigation (SPA)\n var lastUrl = location.href;\n setInterval(function() {\n if (location.href !== lastUrl) {\n lastUrl = location.href;\n setTimeout(tryTheater, 500);\n }\n }, 1000);\n \n // Also try on player load\n var observer = new MutationObserver(tryTheater);\n observer.observe(document.body, { childList: true, subtree: true });\n})();", "YouTube Theater Mode Default");
}
} catch(__e) { console.warn('[Userscript:YouTube Theater Mode Default]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Remove or un-stick sticky/fixed headers that block content\n(function() {\n function unstick() {\n document.querySelectorAll('header, nav, [role=\"banner\"], .header, .navbar, .sticky, .fixed-top, [style*=\"position: fixed\"], [style*=\"position:sticky\"]').forEach(function(el) {\n if (el.style.position === 'fixed' || el.style.position === 'sticky' || \n getComputedStyle(el).position === 'fixed' || getComputedStyle(el).position === 'sticky') {\n el.style.position = 'static';\n el.style.top = 'auto';\n el.style.zIndex = 'auto';\n }\n });\n }\n \n unstick();\n \n var observer = new MutationObserver(unstick);\n observer.observe(document.body, { childList: true, subtree: true, attributes: true, attributeFilter: ['style', 'class'] });\n})();", "Kill Sticky Headers");
}
} catch(__e) { console.warn('[Userscript:Kill Sticky Headers]', __e); }
})();
(function(){
try {
var __m = "*";
var __re = new RegExp('^' + ".*" + '

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down

, 'i');
if (__m === '*' || __re.test(location.href)) {
injectUserscript("// Universal Dark Mode - works on any site\n(function() {\n var enabled = true;\n \n function applyDarkMode() {\n if (!enabled) return;\n \n // Create style element if it doesn't exist\n var style = document.getElementById('universal-dark-mode-style');\n if (!style) {\n style = document.createElement('style');\n style.id = 'universal-dark-mode-style';\n document.head.appendChild(style);\n }\n \n // Dark mode CSS - inverts colors but preserves images/video\n style.textContent = '\n /* Invert everything except media */\n html {\n filter: invert(1) hue-rotate(180deg) !important;\n background: #1a1a2e !important;\n }\n \n /* Restore images, videos, iframes, canvas */\n img, video, iframe, canvas, svg, picture, [style*=\"background-image\"] {\n filter: invert(1) hue-rotate(180deg) !important;\n }\n \n /* Preserve specific elements that should not be inverted */\n .no-dark-mode, .no-dark-mode *,\n [data-theme=\"light\"], [data-theme=\"light\"],\n .ace_editor, .ace_editor *,\n .CodeMirror, .CodeMirror *,\n .monaco-editor, .monaco-editor *,\n .markdown-body pre, .markdown-body pre *,\n .highlight, .highlight *,\n pre code, pre code * {\n filter: none !important;\n }\n \n /* Fix common UI elements */\n .modal, .popup, .dropdown-menu, .tooltip, .popover {\n filter: invert(1) hue-rotate(180deg) !important;\n background: #2d2d44 !important;\n border-color: #444 !important;\n }\n \n /* Scrollbars */\n ::-webkit-scrollbar { background: #1a1a2e !important; }\n ::-webkit-scrollbar-thumb { background: #444 !important; }\n ::-webkit-scrollbar-thumb:hover { background: #555 !important; }\n \n /* Selection */\n ::selection { background: #4ecdc4 !important; color: #1a1a2e !important; }\n ::-moz-selection { background: #4ecdc4 !important; color: #1a1a2e !important; }\n ';\n }\n \n function removeDarkMode() {\n var style = document.getElementById('universal-dark-mode-style');\n if (style) style.remove();\n }\n \n // Toggle with Alt+Shift+D\n document.addEventListener('keydown', function(e) {\n if (e.altKey && e.shiftKey && e.key === 'D') {\n e.preventDefault();\n enabled = !enabled;\n if (enabled) {\n applyDarkMode();\n console.log('[Universal Dark Mode] Enabled');\n } else {\n removeDarkMode();\n console.log('[Universal Dark Mode] Disabled');\n }\n }\n });\n \n // Apply on load\n applyDarkMode();\n \n // Re-apply on dynamic content\n var observer = new MutationObserver(function(mutations) {\n if (enabled && !document.getElementById('universal-dark-mode-style')) {\n applyDarkMode();\n }\n });\n observer.observe(document.head, { childList: true });\n \n console.log('[Universal Dark Mode] Loaded - Press Alt+Shift+D to toggle');\n})();", "Universal Dark Mode");
}
} catch(__e) { console.warn('[Userscript:Universal Dark Mode]', __e); }
})();
})();

Skip to content
Fix unfused QKV params case; stack vs interleave option#83

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

ksivaman
merged 2 commits into

from

Mar 2, 2023
 Merged


Fix unfused QKV params case; stack vs interleave option
#83
Show file tree
Hide file tree
 Changes from all commits
Commits

Show all changes


2 commits


Select commit
Hold shift + click to select a range
b5d4ba6

fix qkv weight unfused path


ksivaman Mar 1, 2023
f561579

fix non FA non interleaved case


ksivaman Mar 1, 2023

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


79 changes: 59 additions & 20 deletions
transformer_engine/pytorch/transformer.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -24,7 +24,7 @@
from transformer_engine.pytorch.utils import (
 divide,
 attention_mask_func,
split_tensor_along_last_dim,
split_tensor_along_dim,
 cast_if_needed,
 get_default_init_method,
)
Expand DownExpand Up@@ -126,11 +126,11 @@ def forward(
 )

 # [sq, b, np, hn] -> [sq, b * np, hn]
 query_layer = query_layer.view(
 query_layer = query_layer.reshape(
 output_size[2], output_size[0] * output_size[1], -1
 )
 # [sk, b, np, hn] -> [sk, b * np, hn]
 key_layer = key_layer.view(output_size[3], output_size[0] * output_size[1], -1)
 key_layer = key_layer.reshape(output_size[3], output_size[0] * output_size[1], -1)

 # preallocting result tensor: [b * np, sq, sk]
 matmul_result = torch.empty(
Expand DownExpand Up@@ -171,7 +171,7 @@ def forward(
 )

 # change view [sk, b * np, hn]
 value_layer = value_layer.view(
 value_layer = value_layer.reshape(
 value_layer.size(0), output_size[0] * output_size[1], -1
 )

Expand DownExpand Up@@ -504,6 +504,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()
 self.layer_number = (layer_number,)
Expand All@@ -515,6 +516,10 @@ def __init__(
 self.params_dtype = params_dtype
 self.init_method = init_method

 if not fuse_qkv_params:
 qkv_weight_interleaved = False
 self.qkv_weight_interleaved = qkv_weight_interleaved

 assert (
 attention_type in AttnTypes
 ), f"attention_type {attention_type} not supported"
Expand DownExpand Up@@ -703,16 +708,28 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 3 * hn)] --> [sq, b, np, 3 * hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 3 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 3 * hn)] --> [sq, b, 3 * np, hn]
 new_tensor_shape = mixed_x_layer.size()[:-1] + (
 3 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_x_layer = mixed_x_layer.view(*new_tensor_shape)

 # [sq, b, np, 3 * hn] --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_last_dim(
 mixed_x_layer, 3
 # mixed_x_layer --> 3 [sq, b, np, hn]
 query_layer, key_layer, value_layer = split_tensor_along_dim(
 mixed_x_layer, split_dim, 3
 )
 else:
 # Attention heads [sk, b, h] --> [sk, b, (np * 2 * hn)]
Expand All@@ -721,15 +738,27 @@ def forward(
 is_first_microbatch=is_first_microbatch,
 )

 # [sk, b, (np * 2 * hn)] --> [sk, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 if self.qkv_weight_interleaved:
 # [sq, b, (np * 2 * hn)] --> [sq, b, np, 2 * hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 self.num_attention_heads_per_partition,
 2 * self.hidden_size_per_attention_head,
 )
 # split along last dimension
 split_dim = -1
 else:
 # [sq, b, (np * 2 * hn)] --> [sq, b, 2 * np, hn]
 new_tensor_shape = mixed_kv_layer.size()[:-1] + (
 2 * self.num_attention_heads_per_partition,
 self.hidden_size_per_attention_head,
 )
 # split along second last dimension
 split_dim = -2

 mixed_kv_layer = mixed_kv_layer.view(*new_tensor_shape)

 # [sk, b, np, 2 * hn] --> 2 [sk, b, np, hn]
(key_layer, value_layer) = split_tensor_along_last_dim(mixed_kv_layer, 2)
 # mixed_kv_layer --> 2 [sk, b, np, hn]
 key_layer, value_layer = split_tensor_along_dim(mixed_kv_layer, split_dim, 2)

 # Attention head [sq, b, h] --> [sq, b, hp]
 if self.input_layernorm:
Expand DownExpand Up@@ -863,7 +892,12 @@ class TransformerLayer(torch.nn.Module):
 .. math::
 y = \frac{x - \mathrm{E}[x]}{ \sqrt{\mathrm{Var}[x] + \varepsilon}} *
 (1 + \gamma) + \beta

 qkv_weight_interleaved : bool, default = `True`
 if set to `False`, the QKV weight is interpreted as a concatenation of
 query, key, and value weights along the `0th` dimension. The default
 interpretation is that the individual `q`, `k`, and `v` weights for each
 attention head are interleaved. This parameter is set to `False` when
 using :attr:`fuse_qkv_params=False`.
 Parallelism parameters
 ----------------------
 set_parallel_mode : bool, default = `False`
Expand DownExpand Up@@ -938,6 +972,7 @@ def __init__(
 set_parallel_mode: bool = False,
 fuse_qkv_params: bool = False,
 zero_centered_gamma: bool = False,
 qkv_weight_interleaved: bool = True,
 ) -> None:
 super().__init__()

Expand All@@ -958,6 +993,9 @@ def __init__(
 not fuse_wgrad_accumulation
 ), "Gradient accumulation fusion requires single QKV parameter."

 if not fuse_qkv_params:
 qkv_weight_interleaved = False

 self.kv_channels = (
 kv_channels if kv_channels else (hidden_size // num_attention_heads)
 )
Expand DownExpand Up@@ -995,6 +1033,7 @@ def __init__(
 "set_parallel_mode": set_parallel_mode,
 "fuse_qkv_params": fuse_qkv_params,
 "zero_centered_gamma": zero_centered_gamma,
 "qkv_weight_interleaved" : qkv_weight_interleaved,
 }

 self.self_attention = MultiHeadAttention(
Expand Down

9 changes: 4 additions & 5 deletions
transformer_engine/pytorch/utils.py

View file


Open in desktop

Original file line numberDiff line numberDiff line change
Expand Up@@ -78,8 +78,8 @@ def divide(numerator: int, denominator: int) -> int:
 return numerator // denominator


def split_tensor_along_last_dim(
 tensor: torch.Tensor, num_partitions: int, contiguous_split_chunks: bool = False
def split_tensor_along_dim(
 tensor: torch.Tensor, dim: int, num_partitions: int, contiguous_split_chunks: bool = False
) -> Tuple[torch.Tensor, ...]:
 """Split a tensor along its last dimension.
 Arguments:
Expand All@@ -89,10 +89,9 @@ def split_tensor_along_last_dim(
 in memory.
 """
 # Get the size and dimension.
 last_dim = tensor.dim() - 1
 last_dim_size = divide(tensor.size()[last_dim], num_partitions)
 split_size = divide(tensor.size()[dim], num_partitions)
 # Split.
 tensor_list = torch.split(tensor, last_dim_size, dim=last_dim)
 tensor_list = torch.split(tensor, split_size, dim=dim)
 # Note: torch.split does not create contiguous tensors by default.
 if contiguous_split_chunks:
 return tuple(chunk.contiguous() for chunk in tensor_list)
Expand Down