diff --git a/doc/DOCKER_IMAGES.md b/doc/DOCKER_IMAGES.md index 47a300eabc85689f9bce7c46c353b35b85db9376..dcaa34b1844bca5347569235fa8309d57f7a6b9c 100644 --- a/doc/DOCKER_IMAGES.md +++ b/doc/DOCKER_IMAGES.md @@ -24,7 +24,6 @@ You can get images in two ways: ``` - ## Image description @@ -40,3 +39,13 @@ Runtime images cannot be used for compilation. | GPU (cuda10.0-cudnn7) development | CentOS7 | latest-cuda10.0-cudnn7-devel | [Dockerfile.cuda10.0-cudnn7.devel](../tools/Dockerfile.cuda10.0-cudnn7.devel) | | CPU development (Used to compile packages on Ubuntu) | CentOS6 | | [Dockerfile.centos6.devel](../tools/Dockerfile.centos6.devel) | | GPU (cuda9.0-cudnn7) development (Used to compile packages on Ubuntu) | CentOS6 | | [Dockerfile.centos6.cuda9.0-cudnn7.devel](../tools/Dockerfile.centos6.cuda9.0-cudnn7.devel) | + + + +## Requirements for running CUDA containers + +Running a CUDA container requires a machine with at least one CUDA-capable GPU and a driver compatible with the CUDA toolkit version you are using. + +The machine running the CUDA container **only requires the NVIDIA driver**, the CUDA toolkit doesn't have to be installed. + +For the relationship between CUDA toolkit version, Driver version and GPU architecture, please refer to [nvidia-docker wiki](https://github.com/NVIDIA/nvidia-docker/wiki/CUDA). diff --git a/doc/DOCKER_IMAGES_CN.md b/doc/DOCKER_IMAGES_CN.md index 26ef5e8bd8c23a281604e5ff0319416c3e408472..6865eb773665f5bc51aa9976b2d330e607e7c83d 100644 --- a/doc/DOCKER_IMAGES_CN.md +++ b/doc/DOCKER_IMAGES_CN.md @@ -22,7 +22,6 @@ ```shell docker build -t : . ``` - @@ -40,3 +39,13 @@ | GPU (cuda10.0-cudnn7) 开发镜像 | CentOS7 | latest-cuda10.0-cudnn7-devel | [Dockerfile.cuda10.0-cudnn7.devel](../tools/Dockerfile.cuda10.0-cudnn7.devel) | | CPU 开发镜像 (用于编译 Ubuntu 包) | CentOS6 | <无> | [Dockerfile.centos6.devel](../tools/Dockerfile.centos6.devel) | | GPU (cuda9.0-cudnn7) 开发镜像 (用于编译 Ubuntu 包) | CentOS6 | <无> | [Dockerfile.centos6.cuda9.0-cudnn7.devel](../tools/Dockerfile.centos6.cuda9.0-cudnn7.devel) | + + + +## 运行CUDA容器的要求 + +运行CUDA容器需要至少具有一个支持CUDA的GPU以及与您所使用的CUDA工具包版本兼容的驱动程序。 + +运行CUDA容器的机器**只需要相应的NVIDIA驱动程序**，而CUDA工具包不是必要的。 + +相关CUDA工具包版本、驱动版本和GPU架构的关系请参阅 [nvidia-docker wiki](https://github.com/NVIDIA/nvidia-docker/wiki/CUDA)。 diff --git a/doc/INFERENCE_TO_SERVING.md b/doc/INFERENCE_TO_SERVING.md index e10ee976fb455c8cc49a0d5fa44ed4cc1f300ba9..719aa63c0a9b408d6bff628e7be4f35dfb49c5c8 100644 --- a/doc/INFERENCE_TO_SERVING.md +++ b/doc/INFERENCE_TO_SERVING.md @@ -24,13 +24,13 @@ inference_model_dir = "your_inference_model" serving_client_dir = "serving_client_dir" serving_server_dir = "serving_server_dir" feed_var_names, fetch_var_names = inference_model_to_serving( - inference_model_dir, serving_client_dir, serving_server_dir) + inference_model_dir, serving_server_dir, serving_client_dir) ``` if your model file and params file are both standalone, please use the following api. ``` feed_var_names, fetch_var_names = inference_model_to_serving( - inference_model_dir, serving_client_dir, serving_server_dir, + inference_model_dir, serving_server_dir, serving_client_dir, model_filename="model", params_filename="params") ``` diff --git a/doc/INFERENCE_TO_SERVING_CN.md b/doc/INFERENCE_TO_SERVING_CN.md index e7e909ac04be3b1a0885b3390d99a153dfbd170e..5d783f25a3f367baa94d471e50f227d9e6f733d1 100644 --- a/doc/INFERENCE_TO_SERVING_CN.md +++ b/doc/INFERENCE_TO_SERVING_CN.md @@ -23,11 +23,11 @@ inference_model_dir = "your_inference_model" serving_client_dir = "serving_client_dir" serving_server_dir = "serving_server_dir" feed_var_names, fetch_var_names = inference_model_to_serving( - inference_model_dir, serving_client_dir, serving_server_dir) + inference_model_dir, serving_server_dir, serving_client_dir) ``` 如果模型中有模型描述文件`model_filename` 和 模型参数文件`params_filename`，那么请用 ``` feed_var_names, fetch_var_names = inference_model_to_serving( - inference_model_dir, serving_client_dir, serving_server_dir, + inference_model_dir, serving_server_dir, serving_client_dir, model_filename="model", params_filename="params") ``` diff --git a/java/examples/pom.xml b/java/examples/pom.xml index b6c8bc424f5d528d74a4a45828fd9b5e7e5d008e..745e8d4f0f3d47e488f99bd7fe73ed6a9f887373 100644 --- a/java/examples/pom.xml +++ b/java/examples/pom.xml @@ -75,7 +75,7 @@ junit junit - 4.11 + 4.13.1 test diff --git a/python/examples/faster_rcnn_model/000000570688.jpg b/python/examples/cascade_rcnn/000000570688.jpg similarity index 100% rename from python/examples/faster_rcnn_model/000000570688.jpg rename to python/examples/cascade_rcnn/000000570688.jpg diff --git a/python/examples/faster_rcnn_model/label_list.txt b/python/examples/cascade_rcnn/label_list.txt similarity index 100% rename from python/examples/faster_rcnn_model/label_list.txt rename to python/examples/cascade_rcnn/label_list.txt diff --git a/python/examples/cascade_rcnn/test_client.py b/python/examples/cascade_rcnn/test_client.py new file mode 100644 index 0000000000000000000000000000000000000000..b40e97acc3e84a7dc7411a7ad3c3f8c1dc8171a6 --- /dev/null +++ b/python/examples/cascade_rcnn/test_client.py @@ -0,0 +1,40 @@ +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. + +from paddle_serving_client import Client +from paddle_serving_app.reader import * +import numpy as np + +preprocess = Sequential([ + File2Image(), BGR2RGB(), Div(255.0), + Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225], False), + Resize(800, 1333), Transpose((2, 0, 1)), PadStride(32) +]) +postprocess = RCNNPostprocess("label_list.txt", "output") +client = Client() +client.load_client_config("serving_client/serving_client_conf.prototxt") +client.connect(['127.0.0.1:9292']) +im = preprocess('000000570688.jpg') +fetch_map = client.predict( + feed={ + "image": im, + "im_info": np.array(list(im.shape[1:]) + [1.0]), + "im_shape": np.array(list(im.shape[1:]) + [1.0]) + }, + fetch=["multiclass_nms_0.tmp_0"], + batch=False) +fetch_map["image"] = '000000570688.jpg' +print(fetch_map) +postprocess(fetch_map) +print(fetch_map) diff --git a/python/examples/faster_rcnn/000000570688.jpg b/python/examples/faster_rcnn/000000570688.jpg new file mode 100644 index 0000000000000000000000000000000000000000..cb304bd56c4010c08611a30dcca58ea9140cea54 Binary files /dev/null and b/python/examples/faster_rcnn/000000570688.jpg differ diff --git a/python/examples/faster_rcnn_model/000000570688_bbox.jpg b/python/examples/faster_rcnn/000000570688_bbox.jpg similarity index 100% rename from python/examples/faster_rcnn_model/000000570688_bbox.jpg rename to python/examples/faster_rcnn/000000570688_bbox.jpg diff --git a/python/examples/faster_rcnn_model/README.md b/python/examples/faster_rcnn/README.md similarity index 100% rename from python/examples/faster_rcnn_model/README.md rename to python/examples/faster_rcnn/README.md diff --git a/python/examples/faster_rcnn_model/README_CN.md b/python/examples/faster_rcnn/README_CN.md similarity index 100% rename from python/examples/faster_rcnn_model/README_CN.md rename to python/examples/faster_rcnn/README_CN.md diff --git a/python/examples/faster_rcnn/benchmark.py b/python/examples/faster_rcnn/benchmark.py new file mode 100755 index 0000000000000000000000000000000000000000..543ef3352ea629f5209b7e5252276bda46a4b2ae --- /dev/null +++ b/python/examples/faster_rcnn/benchmark.py @@ -0,0 +1,123 @@ +# -*- coding: utf-8 -*- +# +# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# pylint: disable=doc-string-missing + +from __future__ import unicode_literals, absolute_import +import os +import sys +import time +import json +import requests +from paddle_serving_client import Client +from paddle_serving_client.utils import MultiThreadRunner +from paddle_serving_client.utils import benchmark_args, show_latency +from paddle_serving_app.reader import ChineseBertReader + +from paddle_serving_app.reader import * +import numpy as np + +args = benchmark_args() + + +def single_func(idx, resource): + img = "./000000570688.jpg" + profile_flags = False + latency_flags = False + if os.getenv("FLAGS_profile_client"): + profile_flags = True + if os.getenv("FLAGS_serving_latency"): + latency_flags = True + latency_list = [] + + if args.request == "rpc": + preprocess = Sequential([ + File2Image(), BGR2RGB(), Div(255.0), + Normalize([0.485, 0.456, 0.406], [0.229, 0.224, 0.225], False), + Resize(640, 640), Transpose((2, 0, 1)) + ]) + + postprocess = RCNNPostprocess("label_list.txt", "output") + client = Client() + + client.load_client_config(args.model) + client.connect([resource["endpoint"][idx % len(resource["endpoint"])]]) + + start = time.time() + for i in range(turns): + if args.batch_size >= 1: + l_start = time.time() + feed_batch = [] + b_start = time.time() + im = preprocess(img) + for bi in range(args.batch_size): + print("1111batch") + print(bi) + feed_batch.append({ + "image": im, + "im_info": np.array(list(im.shape[1:]) + [1.0]), + "im_shape": np.array(list(im.shape[1:]) + [1.0]) + }) + # im = preprocess(img) + b_end = time.time() + + if profile_flags: + sys.stderr.write( + "PROFILE\tpid:{}\tbert_pre_0:{} bert_pre_1:{}\n".format( + os.getpid(), + int(round(b_start * 1000000)), + int(round(b_end * 1000000)))) + #result = client.predict(feed=feed_batch, fetch=fetch) + fetch_map = client.predict( + feed=feed_batch, fetch=["multiclass_nms"]) + fetch_map["image"] = img + postprocess(fetch_map) + + l_end = time.time() + if latency_flags: + latency_list.append(l_end * 1000 - l_start * 1000) + else: + print("unsupport batch size {}".format(args.batch_size)) + else: + raise ValueError("not implemented {} request".format(args.request)) + end = time.time() + if latency_flags: + return [[end - start], latency_list] + else: + return [[end - start]] + + +if __name__ == '__main__': + multi_thread_runner = MultiThreadRunner() + endpoint_list = ["127.0.0.1:7777"] + turns = 10 + start = time.time() + result = multi_thread_runner.run( + single_func, args.thread, {"endpoint": endpoint_list, + "turns": turns}) + end = time.time() + total_cost = end - start + + avg_cost = 0 + for i in range(args.thread): + avg_cost += result[0][i] + avg_cost = avg_cost / args.thread + + print("total cost: {}s".format(total_cost)) + print("each thread cost: {}s. ".format(avg_cost)) + print("qps: {}samples/s".format(args.batch_size * args.thread * turns / + total_cost)) + if os.getenv("FLAGS_serving_latency"): + show_latency(result[1]) diff --git a/python/examples/faster_rcnn/benchmark.sh b/python/examples/faster_rcnn/benchmark.sh new file mode 100755 index 0000000000000000000000000000000000000000..5706fd03c7a0e266bcac18b0544c64f327cbbe9b --- /dev/null +++ b/python/examples/faster_rcnn/benchmark.sh @@ -0,0 +1,52 @@ +rm profile_log* +export CUDA_VISIBLE_DEVICES=0 +export FLAGS_profile_server=1 +export FLAGS_profile_client=1 +export FLAGS_serving_latency=1 + +gpu_id=0 +#save cpu and gpu utilization log +if [ -d utilization ];then + rm -rf utilization +else + mkdir utilization +fi +#start server +$PYTHONROOT/bin/python3 -m paddle_serving_server_gpu.serve --model $1 --port 7777 --thread 4 --gpu_ids 0 --ir_optim > elog 2>&1 & +sleep 5 + +#warm up +$PYTHONROOT/bin/python3 benchmark.py --thread 4 --batch_size 1 --model $2/serving_client_conf.prototxt --request rpc > profile 2>&1 +echo -e "import psutil\ncpu_utilization=psutil.cpu_percent(1,False)\nprint('CPU_UTILIZATION:', cpu_utilization)\n" > cpu_utilization.py +for thread_num in 1 4 8 16 +do +for batch_size in 1 +do + job_bt=`date '+%Y%m%d%H%M%S'` + nvidia-smi --id=0 --query-compute-apps=used_memory --format=csv -lms 100 > gpu_use.log 2>&1 & + nvidia-smi --id=0 --query-gpu=utilization.gpu --format=csv -lms 100 > gpu_utilization.log 2>&1 & + gpu_memory_pid=$! + $PYTHONROOT/bin/python3 benchmark.py --thread $thread_num --batch_size $batch_size --model $2/serving_client_conf.prototxt --request rpc > profile 2>&1 + kill ${gpu_memory_pid} + kill `ps -ef|grep used_memory|awk '{print $2}'` + echo "model_name:" $1 + echo "thread_num:" $thread_num + echo "batch_size:" $batch_size + echo "=================Done====================" + echo "model_name:$1" >> profile_log_$1 + echo "batch_size:$batch_size" >> profile_log_$1 + $PYTHONROOT/bin/python3 cpu_utilization.py >> profile_log_$1 + job_et=`date '+%Y%m%d%H%M%S'` + awk 'BEGIN {max = 0} {if(NR>1){if ($1 > max) max=$1}} END {print "MAX_GPU_MEMORY:", max}' gpu_use.log >> profile_log_$1 + awk 'BEGIN {max = 0} {if(NR>1){if ($1 > max) max=$1}} END {print "GPU_UTILIZATION:", max}' gpu_utilization.log >> profile_log_$1 + rm -rf gpu_use.log gpu_utilization.log + $PYTHONROOT/bin/python3 ../util/show_profile.py profile $thread_num >> profile_log_$1 + tail -n 8 profile >> profile_log_$1 + echo "" >> profile_log_$1 +done +done + +#Divided log +awk 'BEGIN{RS="\n\n"}{i++}{print > "bert_log_"i}' profile_log_$1 +mkdir bert_log && mv bert_log_* bert_log +ps -ef|grep 'serving'|grep -v grep|cut -c 9-15 | xargs kill -9 diff --git a/python/examples/faster_rcnn/label_list.txt b/python/examples/faster_rcnn/label_list.txt new file mode 100644 index 0000000000000000000000000000000000000000..d7d43a94adf73208f997f0efd6581bef11ca734e --- /dev/null +++ b/python/examples/faster_rcnn/label_list.txt @@ -0,0 +1,81 @@ +background +person +bicycle +car +motorcycle +airplane +bus +train +truck +boat +traffic light +fire hydrant +stop sign +parking meter +bench +bird +cat +dog +horse +sheep +cow +elephant +bear +zebra +giraffe +backpack +umbrella +handbag +tie +suitcase +frisbee +skis +snowboard +sports ball +kite +baseball bat +baseball glove +skateboard +surfboard +tennis racket +bottle +wine glass +cup +fork +knife +spoon +bowl +banana +apple +sandwich +orange +broccoli +carrot +hot dog +pizza +donut +cake +chair +couch +potted plant +bed +dining table +toilet +tv +laptop +mouse +remote +keyboard +cell phone +microwave +oven +toaster +sink +refrigerator +book +clock +vase +scissors +teddy bear +hair drier +toothbrush diff --git a/python/examples/faster_rcnn_model/test_client.py b/python/examples/faster_rcnn/test_client.py similarity index 100% rename from python/examples/faster_rcnn_model/test_client.py rename to python/examples/faster_rcnn/test_client.py diff --git a/python/examples/ocr/det_debugger_server.py b/python/examples/ocr/det_debugger_server.py index 913a0bb4c9a099cbef886beb3889337d024d10d6..ebaf0a3066c7e948b4ca37dda06c60423ad28da9 100644 --- a/python/examples/ocr/det_debugger_server.py +++ b/python/examples/ocr/det_debugger_server.py @@ -53,7 +53,9 @@ class OCRService(WebService): self.ori_h, self.ori_w, _ = im.shape det_img = self.det_preprocess(im) _, self.new_h, self.new_w = det_img.shape - return {"image": det_img[np.newaxis, :].copy()}, ["concat_1.tmp_0"] + return { + "image": det_img[np.newaxis, :].copy() + }, ["concat_1.tmp_0"], True def postprocess(self, feed={}, fetch=[], fetch_map=None): det_out = fetch_map["concat_1.tmp_0"] diff --git a/python/examples/ocr/det_web_server.py b/python/examples/ocr/det_web_server.py index 38c6541c70e9871d13dd81751d4edb2bc771a904..e90efc7813e96e6b768a09b4ec9f5108c5878527 100644 --- a/python/examples/ocr/det_web_server.py +++ b/python/examples/ocr/det_web_server.py @@ -54,7 +54,7 @@ class OCRService(WebService): det_img = self.det_preprocess(im) _, self.new_h, self.new_w = det_img.shape print(det_img) - return {"image": det_img}, ["concat_1.tmp_0"] + return {"image": det_img}, ["concat_1.tmp_0"], False def postprocess(self, feed={}, fetch=[], fetch_map=None): det_out = fetch_map["concat_1.tmp_0"] diff --git a/python/examples/ocr/ocr_debugger_server.py b/python/examples/ocr/ocr_debugger_server.py index 3cbc3a66ef620f5c8851b50a352a0c1587467b3b..163ff9788806db819329df769ac44982220f0423 100644 --- a/python/examples/ocr/ocr_debugger_server.py +++ b/python/examples/ocr/ocr_debugger_server.py @@ -42,10 +42,9 @@ class OCRService(WebService): self.det_client = LocalPredictor() if sys.argv[1] == 'gpu': self.det_client.load_model_config( - det_model_config, gpu=True, profile=False) + det_model_config, use_gpu=True, gpu_id=1) elif sys.argv[1] == 'cpu': - self.det_client.load_model_config( - det_model_config, gpu=False, profile=False) + self.det_client.load_model_config(det_model_config) self.ocr_reader = OCRReader() def preprocess(self, feed=[], fetch=[]): @@ -58,7 +57,7 @@ class OCRService(WebService): det_img = det_img[np.newaxis, :] det_img = det_img.copy() det_out = self.det_client.predict( - feed={"image": det_img}, fetch=["concat_1.tmp_0"]) + feed={"image": det_img}, fetch=["concat_1.tmp_0"], batch=True) filter_func = FilterBoxes(10, 10) post_func = DBPostProcess({ "thresh": 0.3, @@ -91,7 +90,7 @@ class OCRService(WebService): imgs[id] = norm_img feed = {"image": imgs.copy()} fetch = ["ctc_greedy_decoder_0.tmp_0", "softmax_0.tmp_0"] - return feed, fetch + return feed, fetch, True def postprocess(self, feed={}, fetch=[], fetch_map=None): rec_res = self.ocr_reader.postprocess(fetch_map, with_score=True) @@ -107,7 +106,8 @@ ocr_service.load_model_config("ocr_rec_model") ocr_service.prepare_server(workdir="workdir", port=9292) ocr_service.init_det_debugger(det_model_config="ocr_det_model") if sys.argv[1] == 'gpu': - ocr_service.run_debugger_service(gpu=True) + ocr_service.set_gpus("2") + ocr_service.run_debugger_service() elif sys.argv[1] == 'cpu': ocr_service.run_debugger_service() ocr_service.run_web_service() diff --git a/python/examples/ocr/ocr_web_client.py b/python/examples/ocr/ocr_web_client.py index 3d25e288dd93014cf9c3f84edc01d42c013ba2d9..ce96a8bbcd585f37368d70070d649e25a0129029 100644 --- a/python/examples/ocr/ocr_web_client.py +++ b/python/examples/ocr/ocr_web_client.py @@ -36,4 +36,5 @@ for img_file in os.listdir(test_img_dir): image = cv2_to_base64(image_data1) data = {"feed": [{"image": image}], "fetch": ["res"]} r = requests.post(url=url, headers=headers, data=json.dumps(data)) + print(r) print(r.json()) diff --git a/python/examples/ocr/ocr_web_server.py b/python/examples/ocr/ocr_web_server.py index de83ca94a4c1f55d886175d9a87b6a34db34c2a5..97e619f2b894144193ee8ed51050f0239616aa39 100644 --- a/python/examples/ocr/ocr_web_server.py +++ b/python/examples/ocr/ocr_web_server.py @@ -50,7 +50,7 @@ class OCRService(WebService): ori_h, ori_w, _ = im.shape det_img = self.det_preprocess(im) det_out = self.det_client.predict( - feed={"image": det_img}, fetch=["concat_1.tmp_0"]) + feed={"image": det_img}, fetch=["concat_1.tmp_0"], batch=False) _, new_h, new_w = det_img.shape filter_func = FilterBoxes(10, 10) post_func = DBPostProcess({ @@ -77,10 +77,10 @@ class OCRService(WebService): max_wh_ratio = max(max_wh_ratio, wh_ratio) for img in img_list: norm_img = self.ocr_reader.resize_norm_img(img, max_wh_ratio) - feed = {"image": norm_img} - feed_list.append(feed) + feed_list.append(norm_img[np.newaxis, :]) + feed_batch = {"image": np.concatenate(feed_list, axis=0)} fetch = ["ctc_greedy_decoder_0.tmp_0", "softmax_0.tmp_0"] - return feed_list, fetch + return feed_batch, fetch, True def postprocess(self, feed={}, fetch=[], fetch_map=None): rec_res = self.ocr_reader.postprocess(fetch_map, with_score=True) diff --git a/python/examples/ocr/rec_debugger_server.py b/python/examples/ocr/rec_debugger_server.py index fbe67aafee5c8dcae269cd4ad6f6100ed514f0b7..9dc7b19b7bf17096cf3a0c18d0a337f990ecd1a4 100644 --- a/python/examples/ocr/rec_debugger_server.py +++ b/python/examples/ocr/rec_debugger_server.py @@ -52,7 +52,7 @@ class OCRService(WebService): imgs[i] = norm_img feed = {"image": imgs.copy()} fetch = ["ctc_greedy_decoder_0.tmp_0", "softmax_0.tmp_0"] - return feed, fetch + return feed, fetch, True def postprocess(self, feed={}, fetch=[], fetch_map=None): rec_res = self.ocr_reader.postprocess(fetch_map, with_score=True) diff --git a/python/examples/ocr/rec_web_server.py b/python/examples/ocr/rec_web_server.py index aae97fd9e3fbd1d29b6cf2ef160b92a522db2e22..300c26be8f6b33c0cdd4a57e75648e444a25d763 100644 --- a/python/examples/ocr/rec_web_server.py +++ b/python/examples/ocr/rec_web_server.py @@ -51,10 +51,17 @@ class OCRService(WebService): max_wh_ratio = max(max_wh_ratio, wh_ratio) for img in img_list: norm_img = self.ocr_reader.resize_norm_img(img, max_wh_ratio) - feed = {"image": norm_img} - feed_list.append(feed) + #feed = {"image": norm_img} + feed_list.append(norm_img) + if len(feed_list) == 1: + feed_batch = { + "image": np.concatenate( + feed_list, axis=0)[np.newaxis, :] + } + else: + feed_batch = {"image": np.concatenate(feed_list, axis=0)} fetch = ["ctc_greedy_decoder_0.tmp_0", "softmax_0.tmp_0"] - return feed_list, fetch + return feed_batch, fetch, True def postprocess(self, feed={}, fetch=[], fetch_map=None): rec_res = self.ocr_reader.postprocess(fetch_map, with_score=True) diff --git a/python/paddle_serving_server/web_service.py b/python/paddle_serving_server/web_service.py index 18e6664edbfd486bb0156ecc58232795f16d74bb..539c613d40af7c1f6824fc03ce3733c942816318 100644 --- a/python/paddle_serving_server/web_service.py +++ b/python/paddle_serving_server/web_service.py @@ -189,7 +189,7 @@ class WebService(object): from paddle_serving_app.local_predict import LocalPredictor self.client = LocalPredictor() self.client.load_model_config( - "{}".format(self.model_config), gpu=False, profile=False) + "{}".format(self.model_config), use_gpu=False) def run_web_service(self): print("This API will be deprecated later. Please do not use it") diff --git a/python/paddle_serving_server_gpu/web_service.py b/python/paddle_serving_server_gpu/web_service.py index 47bf38bf94d6b6444377e3e3967b196bb3edd6a7..c086e393e537a0024d0e605c281fb9344a6ee754 100644 --- a/python/paddle_serving_server_gpu/web_service.py +++ b/python/paddle_serving_server_gpu/web_service.py @@ -250,7 +250,7 @@ class WebService(object): from paddle_serving_app.local_predict import LocalPredictor self.client = LocalPredictor() self.client.load_model_config( - "{}".format(self.model_config), gpu=gpu, profile=False) + "{}".format(self.model_config), use_gpu=True, gpu_id=self.gpus[0]) def run_web_service(self): print("This API will be deprecated later. Please do not use it") diff --git a/python/setup.py.client.in b/python/setup.py.client.in index 527cbbde93b2c2ed0730d8e9c24ec31b9b4ca627..319ae9b24899e99e72f3de8612c44b1ae0d269f2 100644 --- a/python/setup.py.client.in +++ b/python/setup.py.client.in @@ -43,8 +43,8 @@ if '${PACK}' == 'ON': copy_lib() REQUIRED_PACKAGES = [ - 'six >= 1.10.0', 'protobuf >= 3.11.0', 'numpy >= 1.12', 'grpcio >= 1.28.1', - 'grpcio-tools >= 1.28.1' + 'six >= 1.10.0', 'protobuf >= 3.11.0', 'numpy >= 1.12', 'grpcio <= 1.33.2', + 'grpcio-tools <= 1.33.2' ] diff --git a/python/setup.py.server.in b/python/setup.py.server.in index b602f129253e8f9d55ac17175e387f2232182766..ccce8a58480dfe8247a6595deb2b2f823497df0f 100644 --- a/python/setup.py.server.in +++ b/python/setup.py.server.in @@ -28,7 +28,7 @@ max_version, mid_version, min_version = util.python_version() util.gen_pipeline_code("paddle_serving_server") REQUIRED_PACKAGES = [ - 'six >= 1.10.0', 'protobuf >= 3.11.0', 'grpcio >= 1.28.1', 'grpcio-tools >= 1.28.1', + 'six >= 1.10.0', 'protobuf >= 3.11.0', 'grpcio <= 1.33.2', 'grpcio-tools <= 1.33.2', 'paddle_serving_client', 'flask >= 1.1.1', 'paddle_serving_app', 'func_timeout', 'pyyaml' ] diff --git a/python/setup.py.server_gpu.in b/python/setup.py.server_gpu.in index 1303e0404eb9b557dbfb6232ef391aa89c97747a..492df6c6f7d194bc4e94318cf0235f2c2cc7e882 100644 --- a/python/setup.py.server_gpu.in +++ b/python/setup.py.server_gpu.in @@ -30,7 +30,7 @@ max_version, mid_version, min_version = util.python_version() util.gen_pipeline_code("paddle_serving_server_gpu") REQUIRED_PACKAGES = [ - 'six >= 1.10.0', 'protobuf >= 3.11.0', 'grpcio >= 1.28.1', 'grpcio-tools >= 1.28.1', + 'six >= 1.10.0', 'protobuf >= 3.11.0', 'grpcio <= 1.33.2', 'grpcio-tools <= 1.33.2', 'paddle_serving_client', 'flask >= 1.1.1', 'paddle_serving_app', 'func_timeout', 'pyyaml' ] diff --git a/tools/Dockerfile.cuda10.1-cudnn7-trt6.devel b/tools/Dockerfile.cuda10.1-cudnn7-trt6.devel new file mode 100644 index 0000000000000000000000000000000000000000..c6e1c1e050505e631493efe21732a98abd1bd52e --- /dev/null +++ b/tools/Dockerfile.cuda10.1-cudnn7-trt6.devel @@ -0,0 +1,60 @@ +FROM nvidia/cuda:10.1-cudnn7-devel-centos7 + +RUN export http_proxy="http://172.19.56.199:3128" \ + && export https_proxy="http://172.19.56.199:3128" \ + && yum -y install wget >/dev/null \ + && yum -y install gcc gcc-c++ make glibc-static which \ + && yum -y install git openssl-devel curl-devel bzip2-devel python-devel \ + && yum -y install libSM-1.2.2-2.el7.x86_64 --setopt=protected_multilib=false \ + && yum -y install libXrender-0.9.10-1.el7.x86_64 --setopt=protected_multilib=false \ + && yum -y install libXext-1.3.3-3.el7.x86_64 --setopt=protected_multilib=false + +RUN export http_proxy="http://172.19.56.199:3128" \ + && export https_proxy="http://172.19.56.199:3128" && \ + wget https://github.com/protocolbuffers/protobuf/releases/download/v3.11.2/protobuf-all-3.11.2.tar.gz && \ + tar zxf protobuf-all-3.11.2.tar.gz && \ + cd protobuf-3.11.2 && \ + ./configure && make -j4 && make install && \ + make clean && \ + cd .. && rm -rf protobuf-* + +RUN export http_proxy="http://172.19.56.199:3128" \ + && export https_proxy="http://172.19.56.199:3128" && \ + wget https://cmake.org/files/v3.2/cmake-3.2.0-Linux-x86_64.tar.gz >/dev/null \ + && tar xzf cmake-3.2.0-Linux-x86_64.tar.gz \ + && mv cmake-3.2.0-Linux-x86_64 /usr/local/cmake3.2.0 \ + && echo 'export PATH=/usr/local/cmake3.2.0/bin:$PATH' >> /root/.bashrc \ + && rm cmake-3.2.0-Linux-x86_64.tar.gz + + +RUN export http_proxy="http://172.19.56.199:3128" \ + && export https_proxy="http://172.19.56.199:3128" && \ + wget https://dl.google.com/go/go1.14.linux-amd64.tar.gz >/dev/null \ + && tar xzf go1.14.linux-amd64.tar.gz \ + && mv go /usr/local/go \ + && echo 'export GOROOT=/usr/local/go' >> /root/.bashrc \ + && echo 'export PATH=/usr/local/go/bin:$PATH' >> /root/.bashrc \ + && rm go1.14.linux-amd64.tar.gz + +RUN export http_proxy="http://172.19.56.199:3128" \ + && export https_proxy="http://172.19.56.199:3128" && \ + yum -y install python-devel sqlite-devel \ + && curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py >/dev/null \ + && python get-pip.py >/dev/null \ + && rm get-pip.py + +RUN export http_proxy="http://172.19.56.199:3128" \ + && export https_proxy="http://172.19.56.199:3128" && \ + yum install -y python3 python3-devel \ + && yum -y install epel-release && yum -y install patchelf libXext libSM libXrender\ + && yum clean all + +RUN localedef -c -i en_US -f UTF-8 en_US.UTF-8 \ + && echo "export LANG=en_US.utf8" >> /root/.bashrc \ + && echo "export LANGUAGE=en_US.utf8" >> /root/.bashrc + +RUN wget https://paddle-serving.bj.bcebos.com/tools/TensorRT-6.0.1.5.CentOS-7.6.x86_64-gnu.cuda-10.1.cudnn7.6.tar.gz \ + && tar -xzf TensorRT-6.0.1.5.CentOS-7.6.x86_64-gnu.cuda-10.1.cudnn7.6.tar.gz \ + && mv TensorRT-6.0.1.5 /usr/local/ \ + && rm TensorRT-6.0.1.5.CentOS-7.6.x86_64-gnu.cuda-10.1.cudnn7.6.tar.gz \ + && echo 'export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/TensorRT-6.0.1.5/lib/' >> /root/.bashrc