File size: 19,738 Bytes
452b173
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
import numpy as np
import torch
import torch.nn as nn
import math

def quantize(x, scale, zero, maxq):
    if maxq < 0:
        return (x > scale / 2).float() * scale + (x < zero / 2).float() * zero
    q = torch.clamp(torch.round(x / scale) + zero, 0, maxq)
    return scale * (q - zero)

class Quantizer(nn.Module):

    def __init__(self, shape=1):
        super(Quantizer, self).__init__()
        self.register_buffer('maxq', torch.tensor(0))
        self.register_buffer('scale', torch.zeros(shape))
        self.register_buffer('zero', torch.zeros(shape))

    def configure(
        self,
        bits, perchannel=False, sym=True, 
        mse=False, norm=2.4, grid=100, maxshrink=.8,
        trits=False
        ):
        
        self.maxq = torch.tensor(2 ** bits - 1)
        self.perchannel = perchannel
        self.sym = sym
        self.mse = mse
        self.norm = norm
        self.grid = grid
        self.maxshrink = maxshrink 
        if trits:
            self.maxq = torch.tensor(-1) 

    def find_params(self, x, weight=False):
        dev = x.device
        self.maxq = self.maxq.to(dev)

        shape = x.shape
        if self.perchannel:
            if weight:
                x = x.flatten(1)
            else:
                if len(shape) == 4:
                    x = x.permute([1, 0, 2, 3])
                    x = x.flatten(1)
                if len(shape) == 3:
                    x = x.reshape((-1, shape[-1])).t()
                if len(shape) == 2:
                    x = x.t()
        else:
            x = x.flatten().unsqueeze(0)

        tmp = torch.zeros(x.shape[0], device=dev)
        xmin = torch.minimum(x.min(1)[0], tmp)
        xmax = torch.maximum(x.max(1)[0], tmp)

        if self.sym:
            xmax = torch.maximum(torch.abs(xmin), xmax)
            tmp = xmin < 0
            if torch.any(tmp):
                xmin[tmp] = -xmax[tmp]
        tmp = (xmin == 0) & (xmax == 0)
        xmin[tmp] = -1
        xmax[tmp] = +1

        if self.maxq < 0:
            self.scale = xmax
            self.zero = xmin
        else:
            self.scale = (xmax - xmin) / self.maxq
            if self.sym:
                self.zero = torch.full_like(self.scale, (self.maxq + 1) / 2)
            else:
                self.zero = torch.round(-xmin / self.scale)

        if self.mse:
            best = torch.full([x.shape[0]], float('inf'), device=dev)
            for i in range(int(self.maxshrink * self.grid)):
                p = 1 - i / self.grid 
                xmin1 = p * xmin
                xmax1 = p * xmax
                scale1 = (xmax1 - xmin1) / self.maxq
                zero1 = torch.round(-xmin1 / scale1) if not self.sym else self.zero
                q = quantize(x, scale1.unsqueeze(1), zero1.unsqueeze(1), self.maxq)
                q -= x
                q.abs_()
                q.pow_(self.norm)
                err = torch.sum(q, 1)
                tmp = err < best
                if torch.any(tmp):
                    best[tmp] = err[tmp]
                    self.scale[tmp] = scale1[tmp]
                    self.zero[tmp] = zero1[tmp]
        if not self.perchannel:
            if weight:
                tmp = shape[0]
            else:
                tmp = shape[1] if len(shape) != 3 else shape[2]
            self.scale = self.scale.repeat(tmp)
            self.zero = self.zero.repeat(tmp)

        if weight:
            shape = [-1] + [1] * (len(shape) - 1)
            self.scale = self.scale.reshape(shape)
            self.zero = self.zero.reshape(shape)
            return
        if len(shape) == 4:
            self.scale = self.scale.reshape((1, -1, 1, 1))
            self.zero = self.zero.reshape((1, -1, 1, 1))
        if len(shape) == 3:
            self.scale = self.scale.reshape((1, 1, -1))
            self.zero = self.zero.reshape((1, 1, -1)) 
        if len(shape) == 2:
            self.scale = self.scale.unsqueeze(0)
            self.zero = self.zero.unsqueeze(0)

    def quantize(self, x):
        if self.ready():
            return quantize(x, self.scale, self.zero, self.maxq)
        return x

    def enabled(self):
        return self.maxq > 0

    def ready(self):
        return torch.all(self.scale != 0)


try:
    import quant_cuda
except:
    print('CUDA extension not installed.')

# Assumes layer is perfectly divisible into 256 * 256 blocks
class QuantLinear(nn.Module): 
    def __init__(self, bits, groupsize, infeatures, outfeatures, faster=False, kernel_switch_threshold=128):
        super().__init__()
        if bits not in [2,3,4,8]:
            raise NotImplementedError("Only 2,3,4,8 bits are supported.")
        self.infeatures = infeatures
        self.outfeatures = outfeatures
        self.bits = bits
        if groupsize != -1 and groupsize < 32 and groupsize != int(math.pow(2,int(math.log2(groupsize)))):
            raise NotImplementedError("groupsize supports powers of 2 greater than 32. (e.g. : 32,64,128,etc)")
        groupsize = groupsize if groupsize != -1 else infeatures
        self.groupsize = groupsize
        self.register_buffer('qzeros', torch.zeros((math.ceil(infeatures/groupsize),outfeatures // 256 * (bits * 8)), dtype=torch.int))
        self.register_buffer('scales', torch.zeros((math.ceil(infeatures/groupsize),outfeatures)))
        self.register_buffer('bias', torch.zeros(outfeatures))
        self.register_buffer(
            'qweight', torch.zeros((infeatures // 32 * bits, outfeatures), dtype=torch.int)
        )
        self.half_indim = self.infeatures // 2
        self._initialized_quant_state = False
        self.faster = faster
        # kernel_switch_threshold is the cutoff input size after which matmul
        # is performed by unpacking the weights and using torch.matmul
        self.kernel_switch_threshold = kernel_switch_threshold
        if isinstance(self.kernel_switch_threshold, bool):
            self.kernel_switch_threshold = 128 if self.kernel_switch_threshold else None
        if not self.kernel_switch_threshold is None:
            # Buffers for bit shifting weight unpacking
            if self.bits == 2:
                self.register_buffer(
                    'wf1',
                    torch.tensor([0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30], dtype=torch.int32).unsqueeze(0).unsqueeze(2),
                    persistent=False
                )
                self.register_buffer(
                    'wf2',
                    torch.tensor([0, 2, 4, 6, 8, 10, 12, 14, 16, 18, 20, 22, 24, 26, 28, 30], dtype=torch.int32).unsqueeze(0).unsqueeze(0),
                    persistent=False
                )
            elif self.bits == 3:
                self.register_buffer('wf1', torch.tensor([
                    [0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0],
                    [0, 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31],
                    [0, 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0],
                ], dtype=torch.int32).reshape(1,3,12,1), persistent=False)
                self.register_buffer('wf2', torch.tensor([
                    [0, 3, 6, 9, 12, 15, 18, 21, 24, 27, 30, 0],
                    [0, 1, 4, 7, 10, 13, 16, 19, 22, 25, 28, 31],
                    [0, 2, 5, 8, 11, 14, 17, 20, 23, 26, 29, 0],
                ], dtype=torch.int32).reshape(1,1,3,12), persistent=False)
            elif self.bits == 4:
                self.register_buffer(
                    'wf1',
                    torch.tensor([0, 4, 8, 12, 16, 20, 24, 28], dtype=torch.int32).unsqueeze(0).unsqueeze(2),
                    persistent=False
                )
                self.register_buffer(
                    'wf2',
                    torch.tensor([0, 4, 8, 12, 16, 20, 24, 28], dtype=torch.int32).unsqueeze(0).unsqueeze(0),
                    persistent=False
                )
            elif self.bits == 8:
                self.register_buffer(
                    'wf1',
                    torch.tensor([0, 8, 16, 24], dtype=torch.int32).unsqueeze(0).unsqueeze(2),
                    persistent=False
                )
                self.register_buffer(
                    'wf2',
                    torch.tensor([0, 8, 16, 24], dtype=torch.int32).unsqueeze(0).unsqueeze(0),
                    persistent=False
                )
                
    def pack(self, linear, scales, zeros):
        scales = scales.t().contiguous()
        zeros = zeros.t().contiguous()
        scale_zeros = zeros * scales
        self.scales = scales.clone()
        if linear.bias is not None:
            self.bias = linear.bias.clone() 
            
        intweight = []
        for idx in range(self.infeatures):
            g_idx = idx // self.groupsize
            intweight.append(torch.round((linear.weight.data[:,idx] + scale_zeros[g_idx]) / self.scales[g_idx]).to(torch.int)[:,None])
        intweight = torch.cat(intweight,dim=1)
        intweight = intweight.t().contiguous()
        intweight = intweight.numpy().astype(np.uint32)
        qweight = np.zeros(
            (intweight.shape[0] // 32 * self.bits, intweight.shape[1]), dtype=np.uint32
        )
        i = 0
        row = 0
        while row < qweight.shape[0]:
            if self.bits in [2,4,8]:
                for j in range(i, i + (32//self.bits)):
                    qweight[row] |= intweight[j] << (self.bits * (j - i))
                i += 32//self.bits
                row += 1
            elif self.bits == 3:
                for j in range(i, i + 10):
                    qweight[row] |= intweight[j] << (3 * (j - i))
                i += 10
                qweight[row] |= intweight[i] << 30
                row += 1
                qweight[row] |= (intweight[i] >> 2) & 1
                i += 1
                for j in range(i, i + 10):
                    qweight[row] |= intweight[j] << (3 * (j - i) + 1)
                i += 10
                qweight[row] |= intweight[i] << 31
                row += 1
                qweight[row] |= (intweight[i] >> 1) & 0x3
                i += 1
                for j in range(i, i + 10):
                    qweight[row] |= intweight[j] << (3 * (j - i) + 2)
                i += 10
                row += 1
            else:
                raise NotImplementedError("Only 2,3,4,8 bits are supported.")
                
        qweight = qweight.astype(np.int32)
        self.qweight = torch.from_numpy(qweight) 
        
        zeros -= 1;
        zeros = zeros.numpy().astype(np.uint32)
        qzeros = np.zeros((zeros.shape[0], zeros.shape[1] // 256 * (self.bits * 8)), dtype=np.uint32)
        i = 0
        col = 0
        while col < qzeros.shape[1]:
            if self.bits in [2,4,8]:
                for j in range(i, i + (32//self.bits)):
                    qzeros[:, col] |= zeros[:, j] << (self.bits * (j - i))
                i += 32//self.bits
                col += 1
            elif self.bits == 3:
                for j in range(i, i + 10):
                    qzeros[:, col] |= zeros[:, j] << (3 * (j - i))
                i += 10
                qzeros[:, col] |= zeros[:, i] << 30
                col += 1
                qzeros[:, col] |= (zeros[:, i] >> 2) & 1
                i += 1
                for j in range(i, i + 10):
                    qzeros[:, col] |= zeros[:, j] << (3 * (j - i) + 1)
                i += 10
                qzeros[:, col] |= zeros[:, i] << 31
                col += 1
                qzeros[:, col] |= (zeros[:, i] >> 1) & 0x3
                i += 1
                for j in range(i, i + 10):
                    qzeros[:, col] |= zeros[:, j] << (3 * (j - i) + 2)
                i += 10
                col += 1
            else:
                raise NotImplementedError("Only 2,3,4,8 bits are supported.")
                
        qzeros = qzeros.astype(np.int32)
        self.qzeros = torch.from_numpy(qzeros) 

    def forward(self, x):
        if not self._initialized_quant_state:
            # Do we even have a bias? Check for at least one non-zero element.
            if self.bias is not None and bool(torch.any(self.bias != 0)):
                # Then make sure it's the right type.
                self.bias.data = self.bias.data.to(torch.float32)
            else:
                self.bias = None

        if not self.kernel_switch_threshold is None and (x.shape[0] * x.shape[1]) >= self.kernel_switch_threshold:
            if self.bits == 2:
                # Unpack 2bit weights
                
                weight = torch.bitwise_right_shift(torch.unsqueeze(self.qweight, 1).expand(-1, 16, -1), self.wf1).to(torch.int8)
                torch.bitwise_and(weight, 0x00000003, out=weight)
                weight = weight.reshape(-1, self.groupsize, weight.shape[2])

                zeros = torch.bitwise_right_shift(torch.unsqueeze(self.qzeros, 2).expand(-1, -1, 16), self.wf2).to(torch.int8)
                torch.bitwise_and(zeros, 0x00000003, out=zeros)
                zeros = zeros + 1
                zeros = zeros.reshape(-1, 1, zeros.shape[1] * zeros.shape[2])

                scales = self.scales
                scales = scales.reshape(-1, 1, scales.shape[-1])

                weights = (scales * (weight - zeros))
                weights = weights.reshape(weights.shape[0] * weight.shape[1], weights.shape[2])
                x = torch.matmul(x, weights.to(x.dtype))
                x = x + self.bias if self.bias is not None else x
                return x
            
            elif self.bits == 3:
            
                # Unpack 3bit weights
                weight = self.qweight.reshape(self.qweight.shape[0]//3, 3, 1, self.qweight.shape[1]).expand(-1, -1, 12, -1)
                weight = (weight >> self.wf1)&0x7
                weight[:,0,10] = (weight[:,0,10]&0x3) | ((weight[:,1,0] << 2)&0x4)
                weight[:,1,11] = (weight[:,1,11]&0x1) | ((weight[:,2,0] << 1)&0x6)
                weight = weight & 0x7
                weight = torch.cat([weight[:,0,:11], weight[:,1,1:12], weight[:,2,1:11]], dim=1)
                weight = weight.reshape(-1, self.groupsize, weight.shape[2])

                zeros = self.qzeros.reshape(self.qzeros.shape[0], self.qzeros.shape[1]//3, 3, 1).expand(-1, -1, -1, 12)
                zeros = (zeros >> self.wf2)
                zeros[:,:,0,10] = (zeros[:,:,0,10]&0x3) | ((zeros[:,:,1,0] << 2)&0x4)
                zeros[:,:,1,11] = (zeros[:,:,1,11]&0x1) | ((zeros[:,:,2,0] << 1)&0x6)
                zeros = zeros & 0x7
                zeros = torch.cat([zeros[:,:,0,:11], zeros[:,:,1,1:12], zeros[:,:,2,1:11]], dim=2)
                zeros = zeros.reshape(-1, 1, zeros.shape[1] * zeros.shape[2])
                zeros = zeros + 1

                scales = self.scales
                scales = scales.reshape(-1, 1, scales.shape[-1])

                weights = (scales * (weight - zeros))
                weights = weights.reshape(weights.shape[0] * weight.shape[1], weights.shape[2])
                x = torch.matmul(x, weights.to(x.dtype))
                x = x + self.bias if self.bias is not None else x
                return x
                
            elif self.bits == 4:
                # Unpack 4bit weights
                weight = torch.bitwise_right_shift(torch.unsqueeze(self.qweight, 1).expand(-1, 8, -1), self.wf1).to(torch.int8)
                torch.bitwise_and(weight, 0x0000000F, out=weight)
                weight = weight.reshape(-1, self.groupsize, weight.shape[2])

                zeros = torch.bitwise_right_shift(torch.unsqueeze(self.qzeros, 2).expand(-1, -1, 8), self.wf2).to(torch.int8)
                torch.bitwise_and(zeros, 0x0000000F, out=zeros)
                zeros = zeros + 1
                zeros = zeros.reshape(-1, 1, zeros.shape[1] * zeros.shape[2])

                scales = self.scales
                scales = scales.reshape(-1, 1, scales.shape[-1])

                weights = (scales * (weight - zeros))
                weights = weights.reshape(weights.shape[0] * weight.shape[1], weights.shape[2])
                x = torch.matmul(x, weights.to(x.dtype))
                x = x + self.bias if self.bias is not None else x
                return x
                
            elif self.bits == 8:
                # Unpack 8bit weights
                weight = torch.bitwise_right_shift(torch.unsqueeze(self.qweight, 1).expand(-1, 4, -1), self.wf1).to(torch.int8)
                torch.bitwise_and(weight, 0x000000FF, out=weight)
                weight = weight.reshape(-1, self.groupsize, weight.shape[2])

                zeros = torch.bitwise_right_shift(torch.unsqueeze(self.qzeros, 2).expand(-1, -1, 4), self.wf2).to(torch.int8)
                torch.bitwise_and(zeros, 0x000000FF, out=zeros)
                zeros = zeros + 1
                zeros = zeros.reshape(-1, 1, zeros.shape[1] * zeros.shape[2])

                scales = self.scales
                scales = scales.reshape(-1, 1, scales.shape[-1])

                weights = (scales * (weight - zeros))
                weights = weights.reshape(weights.shape[0] * weight.shape[1], weights.shape[2])
                x = torch.matmul(x, weights.to(x.dtype))
                x = x + self.bias if self.bias is not None else x
                return x
            else:
                raise NotImplementedError("Only 2,3,4,8 bits are supported.")

        outshape = list(x.shape)
        outshape[-1] = self.outfeatures
        x = x.reshape(-1, x.shape[-1])
        if self.bias is None:
            y = torch.zeros(x.shape[0], outshape[-1], dtype=torch.float32, device=x.device)
        else:
            y = self.bias.clone().repeat(x.shape[0], 1)

        output_dtype = x.dtype
        if self.faster:
            x = x.half()
            if self.bits == 2:
                quant_cuda.vecquant2matmul_faster(x, self.qweight, y, self.scales, self.qzeros, self.groupsize, self.half_indim)
            elif self.bits == 3:
                quant_cuda.vecquant3matmul_faster(x, self.qweight, y, self.scales, self.qzeros, self.groupsize, self.half_indim)
            elif self.bits == 4:
                quant_cuda.vecquant4matmul_faster(x, self.qweight, y, self.scales, self.qzeros, self.groupsize, self.half_indim)
            else:
                raise NotImplementedError("Only 2,3,4 bits are supported.")
        else:
            x = x.float()
            if self.bits == 2:
                quant_cuda.vecquant2matmul(x, self.qweight, y, self.scales, self.qzeros, self.groupsize)
            elif self.bits == 3:
                quant_cuda.vecquant3matmul(x, self.qweight, y, self.scales, self.qzeros, self.groupsize)
            elif self.bits == 4:
                quant_cuda.vecquant4matmul(x, self.qweight, y, self.scales, self.qzeros, self.groupsize)
            elif self.bits == 8:
                quant_cuda.vecquant8matmul(x, self.qweight, y, self.scales, self.qzeros, self.groupsize)
            else:
                raise NotImplementedError("Only 2,3,4,8 bits are supported.")
        y = y.to(output_dtype)
        return y.reshape(outshape)

def make_quant(module, names, bits, groupsize, faster=False, name='', kernel_switch_threshold=128):
    if isinstance(module, QuantLinear):
        return
    for attr in dir(module):
        tmp = getattr(module, attr)
        name1 = name + '.' + attr if name != '' else attr
        if name1 in names:
            delattr(module, attr)
            setattr(
                module, attr, QuantLinear(bits, groupsize, tmp.in_features, tmp.out_features, faster=faster, kernel_switch_threshold=kernel_switch_threshold)
            )
    for name1, child in module.named_children():
        make_quant(child, names, bits, groupsize, faster, name + '.' + name1 if name != '' else name1, kernel_switch_threshold=kernel_switch_threshold)