-
Notifications
You must be signed in to change notification settings - Fork 2
/
run_QAT.sh
executable file
·226 lines (213 loc) · 7.51 KB
/
run_QAT.sh
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
#! /bin/bash
PRETRAINED_MODEL_PATH="/workspace/pretrained_models"
IMAGENET_DATASET_PATH="/workspace/dataset"
###################################################
GPU_NUM=${1}
# alexnet / resnet20 / resnet50 / densenet121
MODEL=${2}
# svhn / cifar10 / cifar100 / imagenet
DATASET=${3}
BATCH=${4} # 128 / 64 / 32
LEARNING_RATE=${5} # 0.001 / 0.0001
FIRST_RUN=${6} # true / false
CLUSTER=${7} # 16 / 8 / 4 / 2
SUB_CLUSTER=${8} # 32 / 16 / 8 / 4
SIM_METHOD=${9} # and / jaccard
REPR_METHOD="max" # FIXED TO MAX
#####################################################
if [ -z ${CLUSTER} ]; then
if [ "$DATASET" = "imagenet" ]; then
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--imagenet $IMAGENET_DATASET_PATH
elif [ "$MODEL" = resnet20 ]; then
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--torchcv
else
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--dnn_path $PRETRAINED_MODEL_PATH/$DATASET/$MODEL/checkpoint.pth
fi
else
if [ "$FIRST_RUN" = true ]; then
if [ "$DATASET" = "imagenet" ]; then
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--cluster ${CLUSTER} \
--repr_method ${REPR_METHOD} \
--sub_cluster ${SUB_CLUSTER} \
--nnac true \
--similarity_method ${SIM_METHOD} \
--imagenet $IMAGENET_DATASET_PATH
elif [ "$MODEL" = resnet20 ]; then
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--torchcv \
--cluster ${CLUSTER} \
--repr_method ${REPR_METHOD} \
--sub_cluster ${SUB_CLUSTER} \
--nnac true \
--similarity_method ${SIM_METHOD}
else
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--cluster ${CLUSTER} \
--repr_method ${REPR_METHOD} \
--sub_cluster ${SUB_CLUSTER} \
--nnac true \
--similarity_method ${SIM_METHOD} \
--dnn_path $PRETRAINED_MODEL_PATH/$DATASET/$MODEL/checkpoint.pth
fi
else
CLUSTERING_MODEL_PATH="/workspace/PerClusterQuantization/result/kmeans/$MODEL/$DATASET/k${CLUSTER}.part2.${REPR_METHOD}.sub${SUB_CLUSTER}.topk_3.sim_0.7.${SIM_METHOD}/"
if [ "$DATASET" = "imagenet" ]; then
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--clustering_path ${CLUSTERING_MODEL_PATH} \
--cluster ${CLUSTER} \
--repr_method ${REPR_METHOD} \
--sub_cluster ${SUB_CLUSTER} \
--nnac true \
--similarity_method ${SIM_METHOD} \
--imagenet $IMAGENET_DATASET_PATH
elif [ "$MODEL" = resnet20 ]; then
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--torchcv \
--clustering_path ${CLUSTERING_MODEL_PATH} \
--cluster ${CLUSTER} \
--repr_method ${REPR_METHOD} \
--sub_cluster ${SUB_CLUSTER} \
--nnac true \
--similarity_method ${SIM_METHOD}
else
CUDA_VISIBLE_DEVICES=${GPU_NUM} python main.py \
--mode fine \
--epochs 100 \
--batch $BATCH \
--quant_base qat \
--arch $MODEL \
--dataset $DATASET \
--lr $LEARNING_RATE \
--smooth 0.99 \
--bit 4 \
--bit_first 8 \
--bit_classifier 8 \
--bit_addcat 16 \
--per_channel \
--symmetric \
--clustering_path ${CLUSTERING_MODEL_PATH} \
--cluster ${CLUSTER} \
--repr_method ${REPR_METHOD} \
--sub_cluster ${SUB_CLUSTER} \
--nnac true \
--similarity_method ${SIM_METHOD} \
--dnn_path $PRETRAINED_MODEL_PATH/$DATASET/$MODEL/checkpoint.pth
fi
fi
fi