Files
AMD-SHARK-Studio/apps/amdshark_studio/modules/prompt_encoding.py
pdhirajkumarprasad fe03539901 Migration to AMDShark (#2182)
Signed-off-by: pdhirajkumarprasad <dhirajp@amd.com>
2025-11-20 12:52:07 +05:30

377 lines
13 KiB
Python

from typing import List, Optional, Union
from iree import runtime as ireert
import re
import torch
import numpy as np
re_attention = re.compile(
r"""
\\\(|
\\\)|
\\\[|
\\]|
\\\\|
\\|
\(|
\[|
:([+-]?[.\d]+)\)|
\)|
]|
[^\\()\[\]:]+|
:
""",
re.X,
)
def parse_prompt_attention(text):
"""
Parses a string with attention tokens and returns a list of pairs:
text and its associated weight.
Accepted tokens are:
(abc) - increases attention to abc by a multiplier of 1.1
(abc:3.12) - increases attention to abc by a multiplier of 3.12
[abc] - decreases attention to abc by a multiplier of 1.1
\( - literal character '('
\[ - literal character '['
\) - literal character ')'
\] - literal character ']'
\\ - literal character '\'
anything else - just text
>>> parse_prompt_attention('normal text')
[['normal text', 1.0]]
>>> parse_prompt_attention('an (important) word')
[['an ', 1.0], ['important', 1.1], [' word', 1.0]]
>>> parse_prompt_attention('(unbalanced')
[['unbalanced', 1.1]]
>>> parse_prompt_attention('\(literal\]')
[['(literal]', 1.0]]
>>> parse_prompt_attention('(unnecessary)(parens)')
[['unnecessaryparens', 1.1]]
>>> parse_prompt_attention('a (((house:1.3)) [on] a (hill:0.5), sun, (((sky))).')
[['a ', 1.0],
['house', 1.5730000000000004],
[' ', 1.1],
['on', 1.0],
[' a ', 1.1],
['hill', 0.55],
[', sun, ', 1.1],
['sky', 1.4641000000000006],
['.', 1.1]]
"""
res = []
round_brackets = []
square_brackets = []
round_bracket_multiplier = 1.1
square_bracket_multiplier = 1 / 1.1
def multiply_range(start_position, multiplier):
for p in range(start_position, len(res)):
res[p][1] *= multiplier
for m in re_attention.finditer(text):
text = m.group(0)
weight = m.group(1)
if text.startswith("\\"):
res.append([text[1:], 1.0])
elif text == "(":
round_brackets.append(len(res))
elif text == "[":
square_brackets.append(len(res))
elif weight is not None and len(round_brackets) > 0:
multiply_range(round_brackets.pop(), float(weight))
elif text == ")" and len(round_brackets) > 0:
multiply_range(round_brackets.pop(), round_bracket_multiplier)
elif text == "]" and len(square_brackets) > 0:
multiply_range(square_brackets.pop(), square_bracket_multiplier)
else:
res.append([text, 1.0])
for pos in round_brackets:
multiply_range(pos, round_bracket_multiplier)
for pos in square_brackets:
multiply_range(pos, square_bracket_multiplier)
if len(res) == 0:
res = [["", 1.0]]
# merge runs of identical weights
i = 0
while i + 1 < len(res):
if res[i][1] == res[i + 1][1]:
res[i][0] += res[i + 1][0]
res.pop(i + 1)
else:
i += 1
return res
def get_prompts_with_weights(pipe, prompt: List[str], max_length: int):
r"""
Tokenize a list of prompts and return its tokens with weights of each token.
No padding, starting or ending token is included.
"""
tokens = []
weights = []
truncated = False
for text in prompt:
texts_and_weights = parse_prompt_attention(text)
text_token = []
text_weight = []
for word, weight in texts_and_weights:
# tokenize and discard the starting and the ending token
token = pipe.tokenizer(word).input_ids[1:-1]
text_token += token
# copy the weight by length of token
text_weight += [weight] * len(token)
# stop if the text is too long (longer than truncation limit)
if len(text_token) > max_length:
truncated = True
break
# truncate
if len(text_token) > max_length:
truncated = True
text_token = text_token[:max_length]
text_weight = text_weight[:max_length]
tokens.append(text_token)
weights.append(text_weight)
if truncated:
print(
"Prompt was truncated. Try to shorten the prompt or increase max_embeddings_multiples"
)
return tokens, weights
def pad_tokens_and_weights(
tokens,
weights,
max_length,
bos,
eos,
no_boseos_middle=True,
chunk_length=77,
):
r"""
Pad the tokens (with starting and ending tokens) and weights (with 1.0) to max_length.
"""
max_embeddings_multiples = (max_length - 2) // (chunk_length - 2)
weights_length = (
max_length if no_boseos_middle else max_embeddings_multiples * chunk_length
)
for i in range(len(tokens)):
tokens[i] = [bos] + tokens[i] + [eos] * (max_length - 1 - len(tokens[i]))
if no_boseos_middle:
weights[i] = [1.0] + weights[i] + [1.0] * (max_length - 1 - len(weights[i]))
else:
w = []
if len(weights[i]) == 0:
w = [1.0] * weights_length
else:
for j in range(max_embeddings_multiples):
w.append(1.0) # weight for starting token in this chunk
w += weights[i][
j
* (chunk_length - 2) : min(
len(weights[i]), (j + 1) * (chunk_length - 2)
)
]
w.append(1.0) # weight for ending token in this chunk
w += [1.0] * (weights_length - len(w))
weights[i] = w[:]
return tokens, weights
def get_unweighted_text_embeddings(
pipe,
text_input,
chunk_length: int,
no_boseos_middle: Optional[bool] = True,
):
"""
When the length of tokens is a multiple of the capacity of the text encoder,
it should be split into chunks and sent to the text encoder individually.
"""
max_embeddings_multiples = (text_input.shape[1] - 2) // (chunk_length - 2)
if max_embeddings_multiples > 1:
text_embeddings = []
for i in range(max_embeddings_multiples):
# extract the i-th chunk
text_input_chunk = text_input[
:, i * (chunk_length - 2) : (i + 1) * (chunk_length - 2) + 2
].clone()
# cover the head and the tail by the starting and the ending tokens
text_input_chunk[:, 0] = text_input[0, 0]
text_input_chunk[:, -1] = text_input[0, -1]
text_embedding = pipe.run("clip", text_input_chunk)[0].to_host()
if no_boseos_middle:
if i == 0:
# discard the ending token
text_embedding = text_embedding[:, :-1]
elif i == max_embeddings_multiples - 1:
# discard the starting token
text_embedding = text_embedding[:, 1:]
else:
# discard both starting and ending tokens
text_embedding = text_embedding[:, 1:-1]
text_embeddings.append(text_embedding)
# AMDSHARK: Convert the result to tensor
# text_embeddings = torch.concat(text_embeddings, axis=1)
text_embeddings_np = np.concatenate(np.array(text_embeddings))
text_embeddings = torch.from_numpy(text_embeddings_np)
else:
text_embeddings = pipe.run("clip", text_input)[0]
text_embeddings = torch.from_numpy(text_embeddings.to_host())
return text_embeddings
# This function deals with NoneType values occuring in tokens after padding
# It switches out None with 49407 as truncating None values causes matrix dimension errors,
def filter_nonetype_tokens(tokens: List[List]):
return [[49407 if token is None else token for token in tokens[0]]]
def get_weighted_text_embeddings(
pipe,
prompt: List[str],
uncond_prompt: List[str] = None,
max_embeddings_multiples: Optional[int] = 8,
no_boseos_middle: Optional[bool] = True,
skip_parsing: Optional[bool] = False,
skip_weighting: Optional[bool] = False,
):
max_length = (pipe.model_max_length - 2) * max_embeddings_multiples + 2
if not skip_parsing:
prompt_tokens, prompt_weights = get_prompts_with_weights(
pipe, prompt, max_length - 2
)
if uncond_prompt is not None:
uncond_tokens, uncond_weights = get_prompts_with_weights(
pipe, uncond_prompt, max_length - 2
)
else:
prompt_tokens = [
token[1:-1]
for token in pipe.tokenizer(
prompt, max_length=max_length, truncation=True
).input_ids
]
prompt_weights = [[1.0] * len(token) for token in prompt_tokens]
if uncond_prompt is not None:
if isinstance(uncond_prompt, str):
uncond_prompt = [uncond_prompt]
uncond_tokens = [
token[1:-1]
for token in pipe.tokenizer(
uncond_prompt, max_length=max_length, truncation=True
).input_ids
]
uncond_weights = [[1.0] * len(token) for token in uncond_tokens]
# round up the longest length of tokens to a multiple of (model_max_length - 2)
max_length = max([len(token) for token in prompt_tokens])
if uncond_prompt is not None:
max_length = max(max_length, max([len(token) for token in uncond_tokens]))
max_embeddings_multiples = min(
max_embeddings_multiples,
(max_length - 1) // (pipe.model_max_length - 2) + 1,
)
max_embeddings_multiples = max(1, max_embeddings_multiples)
max_length = (pipe.model_max_length - 2) * max_embeddings_multiples + 2
# pad the length of tokens and weights
bos = pipe.tokenizer.bos_token_id
eos = pipe.tokenizer.eos_token_id
prompt_tokens, prompt_weights = pad_tokens_and_weights(
prompt_tokens,
prompt_weights,
max_length,
bos,
eos,
no_boseos_middle=no_boseos_middle,
chunk_length=pipe.model_max_length,
)
# FIXME: This is a hacky fix caused by tokenizer padding with None values
prompt_tokens = filter_nonetype_tokens(prompt_tokens)
# prompt_tokens = torch.tensor(prompt_tokens, dtype=torch.long, device=pipe.device)
prompt_tokens = torch.tensor(prompt_tokens, dtype=torch.long, device="cpu")
if uncond_prompt is not None:
uncond_tokens, uncond_weights = pad_tokens_and_weights(
uncond_tokens,
uncond_weights,
max_length,
bos,
eos,
no_boseos_middle=no_boseos_middle,
chunk_length=pipe.model_max_length,
)
# FIXME: This is a hacky fix caused by tokenizer padding with None values
uncond_tokens = filter_nonetype_tokens(uncond_tokens)
# uncond_tokens = torch.tensor(uncond_tokens, dtype=torch.long, device=pipe.device)
uncond_tokens = torch.tensor(uncond_tokens, dtype=torch.long, device="cpu")
# get the embeddings
text_embeddings = get_unweighted_text_embeddings(
pipe,
prompt_tokens,
pipe.model_max_length,
no_boseos_middle=no_boseos_middle,
)
# prompt_weights = torch.tensor(prompt_weights, dtype=text_embeddings.dtype, device=pipe.device)
prompt_weights = torch.tensor(prompt_weights, dtype=torch.float, device="cpu")
if uncond_prompt is not None:
uncond_embeddings = get_unweighted_text_embeddings(
pipe,
uncond_tokens,
pipe.model_max_length,
no_boseos_middle=no_boseos_middle,
)
# uncond_weights = torch.tensor(uncond_weights, dtype=uncond_embeddings.dtype, device=pipe.device)
uncond_weights = torch.tensor(uncond_weights, dtype=torch.float, device="cpu")
# assign weights to the prompts and normalize in the sense of mean
# TODO: should we normalize by chunk or in a whole (current implementation)?
if (not skip_parsing) and (not skip_weighting):
previous_mean = (
text_embeddings.float().mean(axis=[-2, -1]).to(text_embeddings.dtype)
)
text_embeddings *= prompt_weights.unsqueeze(-1)
current_mean = (
text_embeddings.float().mean(axis=[-2, -1]).to(text_embeddings.dtype)
)
text_embeddings *= (previous_mean / current_mean).unsqueeze(-1).unsqueeze(-1)
if uncond_prompt is not None:
previous_mean = (
uncond_embeddings.float()
.mean(axis=[-2, -1])
.to(uncond_embeddings.dtype)
)
uncond_embeddings *= uncond_weights.unsqueeze(-1)
current_mean = (
uncond_embeddings.float()
.mean(axis=[-2, -1])
.to(uncond_embeddings.dtype)
)
uncond_embeddings *= (
(previous_mean / current_mean).unsqueeze(-1).unsqueeze(-1)
)
if uncond_prompt is not None:
return text_embeddings, uncond_embeddings
return text_embeddings, None