forked from PaddlePaddle/FastDeploy
-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathunet_quant.yaml
More file actions
38 lines (32 loc) · 757 Bytes
/
Copy pathunet_quant.yaml
File metadata and controls
38 lines (32 loc) · 757 Bytes
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
Global:
model_dir: ./Unet_cityscapes_without_argmax_infer/
format: paddle
model_filename: model.pdmodel
params_filename: model.pdiparams
qat_image_path: ./train_stuttgart
ptq_image_path: ./train_stuttgart
input_list: ['x']
qat_preprocess: ppseg_cityscapes_qat_preprocess
ptq_preprocess: ppseg_cityscapes_ptq_preprocess
qat_batch_size: 2
Distillation:
alpha: 1.0
loss: l2
node:
- conv2d_37.tmp_1
QuantAware:
onnx_format: True
quantize_op_types:
- conv2d
- depthwise_conv2d
TrainConfig:
epochs: 10
eval_iter: 180
learning_rate: 0.0005
optimizer_builder:
optimizer:
type: SGD
weight_decay: 4.0e-05
PTQ:
calibration_method: 'avg' # option: avg, abs_max, hist, KL, mse
skip_tensor_list: None