compressed-tensors 0.9.5a20250507__py3-none-any.whl → 0.9.5a20250509__py3-none-any.whl
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- compressed_tensors/compressors/quantized_compressors/__init__.py +1 -0
- compressed_tensors/compressors/quantized_compressors/nvfp4_quantized.py +190 -0
- compressed_tensors/config/base.py +1 -0
- compressed_tensors/version.py +1 -1
- {compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/METADATA +1 -1
- {compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/RECORD +9 -8
- {compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/WHEEL +1 -1
- {compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/licenses/LICENSE +0 -0
- {compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/top_level.txt +0 -0
@@ -0,0 +1,190 @@
|
|
1
|
+
# Copyright (c) 2021 - present / Neuralmagic, Inc. All Rights Reserved.
|
2
|
+
#
|
3
|
+
# Licensed under the Apache License, Version 2.0 (the "License");
|
4
|
+
# you may not use this file except in compliance with the License.
|
5
|
+
# You may obtain a copy of the License at
|
6
|
+
#
|
7
|
+
# http://www.apache.org/licenses/LICENSE-2.0
|
8
|
+
#
|
9
|
+
# Unless required by applicable law or agreed to in writing,
|
10
|
+
# software distributed under the License is distributed on an "AS IS" BASIS,
|
11
|
+
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
12
|
+
# See the License for the specific language governing permissions and
|
13
|
+
# limitations under the License.
|
14
|
+
|
15
|
+
|
16
|
+
from typing import Dict, Optional, Tuple
|
17
|
+
|
18
|
+
import numpy
|
19
|
+
import torch
|
20
|
+
from compressed_tensors.compressors.base import BaseCompressor
|
21
|
+
from compressed_tensors.compressors.quantized_compressors.base import (
|
22
|
+
BaseQuantizationCompressor,
|
23
|
+
)
|
24
|
+
from compressed_tensors.config import CompressionFormat
|
25
|
+
from compressed_tensors.quantization import QuantizationArgs
|
26
|
+
from compressed_tensors.quantization.lifecycle.forward import dequantize, quantize
|
27
|
+
from torch import Tensor
|
28
|
+
|
29
|
+
|
30
|
+
__all__ = ["pack_fp4_to_uint8", "unpack_fp4_from_uint8"]
|
31
|
+
|
32
|
+
FLOAT_TO_E2M1 = [
|
33
|
+
0.0,
|
34
|
+
0.5,
|
35
|
+
1.0,
|
36
|
+
1.5,
|
37
|
+
2.0,
|
38
|
+
3.0,
|
39
|
+
4.0,
|
40
|
+
6.0,
|
41
|
+
]
|
42
|
+
|
43
|
+
|
44
|
+
@BaseCompressor.register(name=CompressionFormat.nvfp4_pack_quantized.value)
|
45
|
+
class NVFP4PackedCompressor(BaseQuantizationCompressor):
|
46
|
+
"""
|
47
|
+
Implements compression of FP4 values. Weights of each quantized layer
|
48
|
+
are packed into uint8. Only supports symmetric weight compression for now.
|
49
|
+
"""
|
50
|
+
|
51
|
+
@property
|
52
|
+
def compression_param_names(self) -> Tuple[str]:
|
53
|
+
"""
|
54
|
+
Returns a tuple of compression parameter names introduced by
|
55
|
+
the compressor during compression
|
56
|
+
"""
|
57
|
+
return (
|
58
|
+
"weight_packed",
|
59
|
+
"weight_scale",
|
60
|
+
"weight_zero_point",
|
61
|
+
"weight_global_scale",
|
62
|
+
)
|
63
|
+
|
64
|
+
def compress_weight(
|
65
|
+
self,
|
66
|
+
weight: Tensor,
|
67
|
+
scale: Tensor,
|
68
|
+
global_scale: Tensor,
|
69
|
+
quantization_args: QuantizationArgs,
|
70
|
+
device: Optional[torch.device] = None,
|
71
|
+
zero_point: Optional[torch.Tensor] = None,
|
72
|
+
g_idx: Optional[torch.Tensor] = None,
|
73
|
+
) -> Dict[str, torch.Tensor]:
|
74
|
+
|
75
|
+
quantized_weight = quantize(
|
76
|
+
x=weight,
|
77
|
+
scale=scale,
|
78
|
+
global_scale=global_scale,
|
79
|
+
zero_point=zero_point,
|
80
|
+
args=quantization_args,
|
81
|
+
)
|
82
|
+
compressed_dict = {}
|
83
|
+
weight_packed = pack_fp4_to_uint8(quantized_weight)
|
84
|
+
if device is not None:
|
85
|
+
weight_packed = weight_packed.to(device)
|
86
|
+
compressed_dict["weight_packed"] = weight_packed
|
87
|
+
return compressed_dict
|
88
|
+
|
89
|
+
def decompress_weight(
|
90
|
+
self,
|
91
|
+
compressed_data: Dict[str, Tensor],
|
92
|
+
quantization_args: Optional[QuantizationArgs] = None,
|
93
|
+
) -> torch.Tensor:
|
94
|
+
|
95
|
+
weight = compressed_data["weight_packed"]
|
96
|
+
scale = compressed_data["weight_scale"]
|
97
|
+
global_scale = compressed_data["weight_global_scale"]
|
98
|
+
m, n = weight.shape
|
99
|
+
# TODO: use a user provided dequant dtype
|
100
|
+
unpacked = unpack_fp4_from_uint8(weight, m, n * 2)
|
101
|
+
decompressed_weight = dequantize(
|
102
|
+
x_q=unpacked, scale=scale, global_scale=global_scale, dtype=unpacked.dtype
|
103
|
+
)
|
104
|
+
|
105
|
+
return decompressed_weight
|
106
|
+
|
107
|
+
|
108
|
+
def pack_fp4_to_uint8(x: torch.Tensor) -> torch.Tensor:
|
109
|
+
"""
|
110
|
+
Packs a tensor with values in the fp4 range into uint8.
|
111
|
+
As there are 16 valid fp4 values, two fp4 values can be
|
112
|
+
packed into one uint8. Each fp4 value is mapped to its
|
113
|
+
particular index (e.g. 0.5 is mapped to index 1, 6.0 is mapped
|
114
|
+
to index 7) which is then represented using 4 bits. Consecutive
|
115
|
+
pairs of 4 bits are then packed into an uint8.
|
116
|
+
|
117
|
+
:param x: tensor to pack
|
118
|
+
returns: a packed tensor in uint8
|
119
|
+
"""
|
120
|
+
|
121
|
+
m, n = x.shape
|
122
|
+
device = x.device
|
123
|
+
|
124
|
+
# Create lookup table for FP4 values to indices
|
125
|
+
# Map the absolute values to 0-7 indices
|
126
|
+
kE2M1 = torch.tensor(FLOAT_TO_E2M1, device=device, dtype=x.dtype)
|
127
|
+
|
128
|
+
# Find closest valid FP4 value index for each element
|
129
|
+
abs_x = torch.abs(x)
|
130
|
+
abs_indices = torch.zeros_like(abs_x, dtype=torch.long)
|
131
|
+
for i, val in enumerate(kE2M1):
|
132
|
+
abs_indices = torch.where(torch.isclose(abs_x, val), i, abs_indices)
|
133
|
+
|
134
|
+
# Apply sign bit (bit 3) to get final 4-bit representation
|
135
|
+
indices = abs_indices + (torch.signbit(x) << 3).to(torch.long)
|
136
|
+
|
137
|
+
# Reshape to prepare for packing pairs of values
|
138
|
+
indices = indices.reshape(-1)
|
139
|
+
|
140
|
+
# Handle odd length by padding if necessary
|
141
|
+
if indices.numel() % 2 != 0:
|
142
|
+
indices = torch.cat([indices, torch.zeros(1, dtype=torch.long, device=device)])
|
143
|
+
|
144
|
+
# Reshape to pair consecutive elements
|
145
|
+
indices = indices.reshape(-1, 2)
|
146
|
+
|
147
|
+
# Pack pairs of 4-bit values into 8-bit values
|
148
|
+
packed = (indices[:, 0] | (indices[:, 1] << 4)).to(torch.uint8)
|
149
|
+
|
150
|
+
return packed.reshape(m, n // 2)
|
151
|
+
|
152
|
+
|
153
|
+
kE2M1ToFloat = torch.tensor(
|
154
|
+
[0.0, 0.5, 1.0, 1.5, 2.0, 3.0, 4.0, 6.0], dtype=torch.float32
|
155
|
+
)
|
156
|
+
|
157
|
+
# reference: : https://github.com/vllm-project/vllm/pull/16362
|
158
|
+
def unpack_fp4_from_uint8(
|
159
|
+
a: torch.Tensor, m: int, n: int, dtype: Optional[torch.dtype] = torch.bfloat16
|
160
|
+
) -> torch.Tensor:
|
161
|
+
"""
|
162
|
+
Unpacks uint8 values into fp4. Each uint8 consists of two fp4 values
|
163
|
+
(i.e. first four bits correspond to one fp4 value, last four corresond to a consecutive
|
164
|
+
fp4 value). The bits represent an index, which are mapped to an fp4 value.
|
165
|
+
|
166
|
+
:param a: tensor to unpack
|
167
|
+
:param m: original dim 0 size of the unpacked tensor
|
168
|
+
:param n: original dim 1 size of the unpacked tensor
|
169
|
+
:param dtype: dense dtype to cast the unpacked tensor to
|
170
|
+
"""
|
171
|
+
assert a.dtype == torch.uint8
|
172
|
+
|
173
|
+
# Vectorized nibble processing
|
174
|
+
a_flat = a.flatten()
|
175
|
+
high = (a_flat & 0xF0) >> 4 # Upper nibbles
|
176
|
+
low = a_flat & 0x0F # Lower nibbles
|
177
|
+
|
178
|
+
# Combine nibbles for batch processing
|
179
|
+
combined = torch.stack((low, high), dim=1).flatten()
|
180
|
+
|
181
|
+
# Vectorized sign and magnitude extraction
|
182
|
+
signs = (combined & 0x08).to(torch.bool) # Sign bits
|
183
|
+
abs_vals = (combined & 0x07).to(torch.long) # Magnitude indices
|
184
|
+
|
185
|
+
# Device-aware lookup and sign application
|
186
|
+
kE2M1 = kE2M1ToFloat.to(device=a.device)
|
187
|
+
values = kE2M1[abs_vals] * torch.where(signs, -1.0, 1.0)
|
188
|
+
|
189
|
+
# Reshape to final form
|
190
|
+
return values.reshape(m, n).to(dtype=dtype)
|
compressed_tensors/version.py
CHANGED
{compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/METADATA
RENAMED
@@ -1,6 +1,6 @@
|
|
1
1
|
Metadata-Version: 2.4
|
2
2
|
Name: compressed-tensors
|
3
|
-
Version: 0.9.
|
3
|
+
Version: 0.9.5a20250509
|
4
4
|
Summary: Library for utilization of compressed safetensors of neural network models
|
5
5
|
Home-page: https://github.com/neuralmagic/compressed-tensors
|
6
6
|
Author: Neuralmagic, Inc.
|
{compressed_tensors-0.9.5a20250507.dist-info → compressed_tensors-0.9.5a20250509.dist-info}/RECORD
RENAMED
@@ -1,14 +1,15 @@
|
|
1
1
|
compressed_tensors/__init__.py,sha256=UtKmifNeBCSE2TZSAfduVNNzHY-3V7bLjZ7n7RuXLOE,812
|
2
2
|
compressed_tensors/base.py,sha256=73HYH7HY7O2roC89yG_piPFnZwrBfn_i7HmKl90SKc0,875
|
3
|
-
compressed_tensors/version.py,sha256=
|
3
|
+
compressed_tensors/version.py,sha256=MJNvDWAetuDB0OYncH2iIZtFhxnO8XXV5IHnTcj9e6k,521
|
4
4
|
compressed_tensors/compressors/__init__.py,sha256=smSygTSfcfuujRrAXDc6uZm4L_ccV1tWZewqVnOb4lM,825
|
5
5
|
compressed_tensors/compressors/base.py,sha256=nvWsv4xEw1Tkxkxth6TmHplDYXfBeP22xWxOsZERyDY,7204
|
6
6
|
compressed_tensors/compressors/helpers.py,sha256=OK6qxX9j3bHwF9JfIYSGMgBJe2PWjlTA3byXKCJaTIQ,5431
|
7
7
|
compressed_tensors/compressors/model_compressors/__init__.py,sha256=5RGGPFu4YqEt_aOdFSQYFYFDjcZFJN0CsMqRtDZz3Js,666
|
8
8
|
compressed_tensors/compressors/model_compressors/model_compressor.py,sha256=uh3Rbyqhjvt8o8On6ioOn6utBKv2siRRmAvgM1lDrxU,26555
|
9
|
-
compressed_tensors/compressors/quantized_compressors/__init__.py,sha256=
|
9
|
+
compressed_tensors/compressors/quantized_compressors/__init__.py,sha256=KvaFBL_Q84LxRGJOV035M8OBoCkAx8kOkfphswgkKWk,745
|
10
10
|
compressed_tensors/compressors/quantized_compressors/base.py,sha256=n0L2QH2_Y1vWtLeQ0uV78y2lV4bviFEAtUKODl8L_nw,8828
|
11
11
|
compressed_tensors/compressors/quantized_compressors/naive_quantized.py,sha256=fd0KlkSx6bvZ3xwIkK3jEUdPSUPs56Eua4dEDOtzKW0,5150
|
12
|
+
compressed_tensors/compressors/quantized_compressors/nvfp4_quantized.py,sha256=Gw-lVzk5jrKUlM5UTCiJBmhM5gHzB9mn8r298MVUbDI,6395
|
12
13
|
compressed_tensors/compressors/quantized_compressors/pack_quantized.py,sha256=SPIHlk8ewip2LcjgkCw02K21EkfUSFSd9qQqL0Pt5eM,11162
|
13
14
|
compressed_tensors/compressors/sparse_compressors/__init__.py,sha256=Atuz-OdEgn8OCUhx7Ovd6gXdyImAI186uCR-uR0t_Nk,737
|
14
15
|
compressed_tensors/compressors/sparse_compressors/base.py,sha256=PMiWIaW2XSF_esYJlQ12RVW7opeAzavdbkRFtelMFX0,6655
|
@@ -18,7 +19,7 @@ compressed_tensors/compressors/sparse_compressors/sparse_bitmask.py,sha256=S8vW0
|
|
18
19
|
compressed_tensors/compressors/sparse_quantized_compressors/__init__.py,sha256=4f_cwcKXB1nVVMoiKgTFAc8jAPjPLElo-Df_EDm1_xw,675
|
19
20
|
compressed_tensors/compressors/sparse_quantized_compressors/marlin_24.py,sha256=7VRLmtUTg1iJl3mXiOzLPi1RgIOhMISPAwzVi8v2QF0,9951
|
20
21
|
compressed_tensors/config/__init__.py,sha256=8sOoZ6xvYSC79mBvEtO8l6xk4PC80d29AnnJiGMrY2M,737
|
21
|
-
compressed_tensors/config/base.py,sha256=
|
22
|
+
compressed_tensors/config/base.py,sha256=p3glQHvC2fjodf_SvlelVrTWSIjGXgGC86t8oVOlMng,3529
|
22
23
|
compressed_tensors/config/dense.py,sha256=NgSxnFCnckU9-iunxEaqiFwqgdO7YYxlWKR74jNbjks,1317
|
23
24
|
compressed_tensors/config/sparse_24_bitmask.py,sha256=Lhj39zT2V1hxftprvxvneyhv45ShlXOKd75DBbDTyTE,1401
|
24
25
|
compressed_tensors/config/sparse_bitmask.py,sha256=pZUboRNZTu6NajGOQEFExoPknak5ynVAUeiiYpS1Gt8,1308
|
@@ -45,8 +46,8 @@ compressed_tensors/utils/permutations_24.py,sha256=kx6fsfDHebx94zsSzhXGyCyuC9sVy
|
|
45
46
|
compressed_tensors/utils/permute.py,sha256=V6tJLKo3Syccj-viv4F7ZKZgJeCB-hl-dK8RKI_kBwI,2355
|
46
47
|
compressed_tensors/utils/safetensors_load.py,sha256=kkkUDmS1H40MFy6FDP-DFGiAYbtqke6bKE7YrAtORtA,11499
|
47
48
|
compressed_tensors/utils/semi_structured_conversions.py,sha256=XKNffPum54kPASgqKzgKvyeqWPAkair2XEQXjkp7ho8,13489
|
48
|
-
compressed_tensors-0.9.
|
49
|
-
compressed_tensors-0.9.
|
50
|
-
compressed_tensors-0.9.
|
51
|
-
compressed_tensors-0.9.
|
52
|
-
compressed_tensors-0.9.
|
49
|
+
compressed_tensors-0.9.5a20250509.dist-info/licenses/LICENSE,sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ,11357
|
50
|
+
compressed_tensors-0.9.5a20250509.dist-info/METADATA,sha256=2BVor0VJtcZHud4QBjYS5OhLXob0nF9dmeP08RUSA5k,7004
|
51
|
+
compressed_tensors-0.9.5a20250509.dist-info/WHEEL,sha256=DnLRTWE75wApRYVsjgc6wsVswC54sMSJhAEd4xhDpBk,91
|
52
|
+
compressed_tensors-0.9.5a20250509.dist-info/top_level.txt,sha256=w2i-GyPs2s1UwVxvutSvN_lM22SXC2hQFBmoMcPnV7Y,19
|
53
|
+
compressed_tensors-0.9.5a20250509.dist-info/RECORD,,
|
File without changes
|
File without changes
|