File size: 10,335 Bytes
b9425fd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
import torch
from typing import Any, Iterable, List, Tuple, Callable
import torch.distributed as dist

def get_gpu_states(fwd_gpu_devices) -> Tuple[List[int], List[torch.Tensor]]:
    # This will not error out if "arg" is a CPU tensor or a non-tensor type because
    # the conditionals short-circuit.
    fwd_gpu_states = []
    for device in fwd_gpu_devices:
        with torch.cuda.device(device):
            fwd_gpu_states.append(torch.cuda.get_rng_state())

    return fwd_gpu_states

def get_gpu_device(*args):

    fwd_gpu_devices = list(set(arg.get_device() for arg in args
                               if isinstance(arg, torch.Tensor) and arg.is_cuda))
    return fwd_gpu_devices

def set_device_states(fwd_cpu_state, devices, states) -> None:
    torch.set_rng_state(fwd_cpu_state)
    for device, state in zip(devices, states):
        with torch.cuda.device(device):
            torch.cuda.set_rng_state(state)

def detach_and_grad(inputs: Tuple[Any, ...]) -> Tuple[torch.Tensor, ...]:
    if isinstance(inputs, tuple):
        out = []
        for inp in inputs:
            if not isinstance(inp, torch.Tensor):
                out.append(inp)
                continue

            x = inp.detach()
            x.requires_grad = True
            out.append(x)
        return tuple(out)
    else:
        raise RuntimeError(
            "Only tuple of tensors is supported. Got Unsupported input type: ", type(inputs).__name__)

def get_cpu_and_gpu_states(gpu_devices):
    return torch.get_rng_state(), get_gpu_states(gpu_devices)

class ReverseFunction(torch.autograd.Function):
    @staticmethod
    def forward(ctx, run_functions, alpha, *args):
        l0, l1, l2, l3 = run_functions
        alpha0, alpha1, alpha2, alpha3 = alpha
        ctx.run_functions  = run_functions
        ctx.alpha = alpha
        ctx.preserve_rng_state = True

        ctx.gpu_autocast_kwargs = {"enabled": torch.is_autocast_enabled(),
                                   "dtype": torch.get_autocast_gpu_dtype(),
                                   "cache_enabled": torch.is_autocast_cache_enabled()}
        ctx.cpu_autocast_kwargs = {"enabled": torch.is_autocast_cpu_enabled(),
                                   "dtype": torch.get_autocast_cpu_dtype(),
                                   "cache_enabled": torch.is_autocast_cache_enabled()}

        assert len(args) == 5
        [x, c0, c1, c2, c3] = args
        if type(c0) == int:
            ctx.first_col = True
        else:
            ctx.first_col = False
        with torch.no_grad():
            if ctx.preserve_rng_state:
                gpu_devices = get_gpu_device(*args)
                ctx.gpu_devices = gpu_devices
                ctx.cpu_states_0, ctx.gpu_states_0  = get_cpu_and_gpu_states(gpu_devices)
                c0 = l0(x, c1, c3) + c0*alpha0
                ctx.cpu_states_1, ctx.gpu_states_1  = get_cpu_and_gpu_states(gpu_devices)
                c1 = l1(c0, c2) + c1*alpha1
                ctx.cpu_states_2, ctx.gpu_states_2  = get_cpu_and_gpu_states(gpu_devices)
                c2 = l2(c1, c3) + c2*alpha2
                ctx.cpu_states_3, ctx.gpu_states_3  = get_cpu_and_gpu_states(gpu_devices)
                c3 = l3(c2) + c3*alpha3
            else:
                c0 = l0(x, c1, c3) + c0*alpha0
                c1 = l1(c0, c2) + c1*alpha1
                c2 = l2(c1, c3) + c2*alpha2
                c3 = l3(c2) + c3*alpha3
        ctx.save_for_backward(x, c0, c1, c2, c3)
        return x, c0, c1 ,c2, c3
    
    @staticmethod
    def backward(ctx, *grad_outputs):
        x, c0, c1, c2, c3 = ctx.saved_tensors
        l0, l1, l2, l3 = ctx.run_functions
        alpha0, alpha1, alpha2, alpha3 = ctx.alpha
        gx_right, g0_right, g1_right, g2_right, g3_right = grad_outputs
        (x, c0, c1, c2, c3) = detach_and_grad((x, c0, c1, c2, c3))

        if ctx.preserve_rng_state:
            with torch.enable_grad(), \
                torch.random.fork_rng(devices=ctx.gpu_devices, enabled=ctx.preserve_rng_state), \
                torch.cuda.amp.autocast(**ctx.gpu_autocast_kwargs), \
                torch.cpu.amp.autocast(**ctx.cpu_autocast_kwargs):
                
                g3_up = g3_right
                g3_left = g3_up*alpha3 ##shortcut
                set_device_states(ctx.cpu_states_3, ctx.gpu_devices, ctx.gpu_states_3)                    
                oup3 = l3(c2)
                torch.autograd.backward(oup3, g3_up, retain_graph=True)
                with torch.no_grad():
                    c3_left = (1/alpha3)*(c3 - oup3) ## feature reverse
                g2_up = g2_right+ c2.grad
                g2_left = g2_up*alpha2 ##shortcut

                (c3_left,) = detach_and_grad((c3_left,))
                set_device_states(ctx.cpu_states_2, ctx.gpu_devices, ctx.gpu_states_2)          
                oup2 = l2(c1, c3_left)
                torch.autograd.backward(oup2, g2_up, retain_graph=True)
                c3_left.requires_grad = False
                cout3 = c3_left*alpha3 ##alpha3 update
                torch.autograd.backward(cout3, g3_up)
                
                with torch.no_grad():
                    c2_left = (1/alpha2)*(c2 - oup2) ## feature reverse
                g3_left = g3_left + c3_left.grad if c3_left.grad is not None else g3_left
                g1_up = g1_right+c1.grad
                g1_left = g1_up*alpha1 ##shortcut

                (c2_left,) = detach_and_grad((c2_left,))
                set_device_states(ctx.cpu_states_1, ctx.gpu_devices, ctx.gpu_states_1)     
                oup1 = l1(c0, c2_left)
                torch.autograd.backward(oup1, g1_up, retain_graph=True)
                c2_left.requires_grad = False
                cout2 = c2_left*alpha2 ##alpha3 update
                torch.autograd.backward(cout2, g2_up)

                with torch.no_grad():
                    c1_left = (1/alpha1)*(c1 - oup1) ## feature reverse
                g0_up = g0_right + c0.grad
                g0_left = g0_up*alpha0 ##shortcut
                g2_left = g2_left + c2_left.grad if c2_left.grad is not None else g2_left ## Fusion
                
                (c1_left,c3_left) = detach_and_grad((c1_left,c3_left))
                set_device_states(ctx.cpu_states_0, ctx.gpu_devices, ctx.gpu_states_0)     
                oup0 = l0(x, c1_left, c3_left)            
                torch.autograd.backward(oup0, g0_up, retain_graph=True)
                c1_left.requires_grad = False
                cout1 = c1_left*alpha1 ##alpha3 update
                torch.autograd.backward(cout1, g1_up)

                with torch.no_grad():
                    c0_left = (1/alpha0)*(c0 - oup0) ## feature reverse
                gx_up = x.grad ## Fusion
                g1_left = g1_left + c1_left.grad if c1_left.grad is not None else g1_left ## Fusion
                g3_left = g3_left + c3_left.grad if c3_left.grad is not None else g3_left ## Fusion
                c0_left.requires_grad = False
                cout0 = c0_left*alpha0 ##alpha3 update
                torch.autograd.backward(cout0, g0_up)
        else:
            with torch.enable_grad():
                
                g3_up = g3_right
                g3_left = g3_up*alpha3 ##shortcut
                oup3 = l3(c2)
                torch.autograd.backward(oup3, g3_up, retain_graph=True)
                with torch.no_grad():
                    c3_left = (1/alpha3)*(c3 - oup3) ## feature reverse
                g2_up = g2_right+ c2.grad
                g2_left = g2_up*alpha2 ##shortcut

                (c3_left,) = detach_and_grad((c3_left,))
                oup2 = l2(c1, c3_left)
                torch.autograd.backward(oup2, g2_up, retain_graph=True)
                c3_left.requires_grad = False
                cout3 = c3_left*alpha3 ##alpha3 update
                torch.autograd.backward(cout3, g3_up)
                
                with torch.no_grad():
                    c2_left = (1/alpha2)*(c2 - oup2) ## feature reverse
                g3_left = g3_left + c3_left.grad if c3_left.grad is not None else g3_left
                g1_up = g1_right+c1.grad
                g1_left = g1_up*alpha1 ##shortcut

                (c2_left,) = detach_and_grad((c2_left,))
                oup1 = l1(c0, c2_left)
                torch.autograd.backward(oup1, g1_up, retain_graph=True)
                c2_left.requires_grad = False
                cout2 = c2_left*alpha2 ##alpha3 update
                torch.autograd.backward(cout2, g2_up)

                with torch.no_grad():
                    c1_left = (1/alpha1)*(c1 - oup1) ## feature reverse
                g0_up = g0_right + c0.grad
                g0_left = g0_up*alpha0 ##shortcut
                g2_left = g2_left + c2_left.grad if c2_left.grad is not None else g2_left ## Fusion
                
                (c1_left,c3_left) = detach_and_grad((c1_left,c3_left))
                oup0 = l0(x, c1_left, c3_left)            
                torch.autograd.backward(oup0, g0_up, retain_graph=True)
                c1_left.requires_grad = False
                cout1 = c1_left*alpha1 ##alpha3 update
                torch.autograd.backward(cout1, g1_up)

                with torch.no_grad():
                    c0_left = (1/alpha0)*(c0 - oup0) ## feature reverse
                gx_up = x.grad ## Fusion
                g1_left = g1_left + c1_left.grad if c1_left.grad is not None else g1_left ## Fusion
                g3_left = g3_left + c3_left.grad if c3_left.grad is not None else g3_left ## Fusion
                c0_left.requires_grad = False
                cout0 = c0_left*alpha0 ##alpha3 update
                torch.autograd.backward(cout0, g0_up)
        # if dist.get_rank()==0:
        #     print(c0_left.mean().data)
        #     print(f'c0: {c0_left.max()}, c1: {c1_left.max()}, c2: {c2_left.max()}, c3: {c3_left.max()}')
        # print(f'x.grad: {gx_up.mean()}, c0.grad: {g0_left.mean()}, c1.grad: {g1_left.mean()}, c2.grad: {g2_left.mean()}, c3.grad: {g3_left.mean()}')
        # import pdb;pdb.set_trace()

        if ctx.first_col:
            # print(f'c0: {c0_left.max()}, c1: {c1_left.max()}, c2: {c2_left.max()}, c3: {c3_left.max()}')
            return None, None, gx_up, None, None, None, None
        else:
            return None, None, gx_up, g0_left, g1_left, g2_left, g3_left