apps/FastDeploy
1
0
Fork 0
mirror of https://github.com/PaddlePaddle/FastDeploy.git synced 2025-10-06 17:17:14 +08:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

[Backend] Add OCR、Seg、 KeypointDetection、Matting、 ernie-3.0 and adaface models for XPU Deploy (#960)

Browse Source 
* [FlyCV] Bump up FlyCV -> official release 1.0.0

* add seg models for XPU

* add ocr model for XPU

* add matting

* add matting python

* fix infer.cc

* add keypointdetection support for XPU

* Add adaface support for XPU

* add ernie-3.0

* fix doc

Co-authored-by: DefTruth <qiustudent_r@163.com>
Co-authored-by: DefTruth <31974251+DefTruth@users.noreply.github.com>
This commit is contained in:
yeliang2258
2022-12-26 15:02:58 +08:00
committed by GitHub
parent 3b29e86add
commit 7b15f72516
39 changed files with 304 additions and 25 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
examples/text/ernie-3.0/python/README.md Normal file → Executable file
View File

@@ -22,7 +22,7 @@ pip install -r requirements.txt
### A Quick Start
The following example shows how to employ FastDeploy library to complete Python predictive deployment of ERNIE 3.0 Medium model on [AFQMC Dataset](https://bj.bcebos.com/paddlenlp/datasets/afqmc_public.zip)of CLUE Benchmark for text classification tasks.
The following example shows how to employ FastDeploy library to complete Python predictive deployment of ERNIE 3.0 Medium model on [AFQMC Dataset](https://bj.bcebos.com/paddlenlp/datasets/afqmc_public.zip)of CLUE Benchmark for text classification tasks.
```bash
@@ -40,6 +40,9 @@ python seq_cls_infer.py --device cpu --model_dir ernie-3.0-medium-zh-afqmc
# GPU Inference
python seq_cls_infer.py --device gpu --model_dir ernie-3.0-medium-zh-afqmc
# XPU Inference
python seq_cls_infer.py --device xpu --model_dir ernie-3.0-medium-zh-afqmc
```
The result returned after running is as follows:

8
examples/text/ernie-3.0/python/seq_cls_infer.py Normal file → Executable file
View File

@@ -35,8 +35,8 @@ def parse_arguments():
"--device",
type=str,
default='cpu',
choices=['gpu', 'cpu'],
help="Type of inference device, support 'cpu' or 'gpu'.")
choices=['gpu', 'cpu', 'xpu'],
help="Type of inference device, support 'cpu', 'xpu' or 'gpu'.")
parser.add_argument(
"--backend",
type=str,
@@ -94,6 +94,10 @@ class ErnieForSequenceClassificationPredictor(object):
model_path = os.path.join(args.model_dir, "infer.pdmodel")
params_path = os.path.join(args.model_dir, "infer.pdiparams")
option.set_model_path(model_path, params_path)
if args.device == 'xpu':
option.use_xpu()
option.use_paddle_lite_backend()
return fd.Runtime(option)
if args.device == 'cpu':
option.use_cpu()
else: