diff --git a/app/configs/global_var.py b/app/configs/global_var.py new file mode 100644 index 0000000..bb2add3 --- /dev/null +++ b/app/configs/global_var.py @@ -0,0 +1,27 @@ +""" +@Time ： 2022/11/15 10:13 +@Auth ： 东 +@File ：global_var.py +@IDE ：PyCharm +@Motto：ABC(Always Be Coding) +@Desc： + +""" + + +def _init(): # 初始化 + global _global_dict + _global_dict = {} + + +def set_value(key, value): + # 定义一个全局变量 + _global_dict[key] = value + + +def get_value(key): + # 获得一个全局变量，不存在则提示读取对应变量失败 + try: + return _global_dict[key] + except: + print('读取' + key + '失败\r\n') diff --git a/app/controller/AlgorithmController.py b/app/controller/AlgorithmController.py index 75d7c38..cc79d94 100644 --- a/app/controller/AlgorithmController.py +++ b/app/controller/AlgorithmController.py @@ -31,6 +31,7 @@ from pathlib import Path bp = Blueprint('AlgorithmController', __name__) +ifKillDict = {} def start_train_algorithm(): """ @@ -147,6 +148,42 @@ def algorithm_process_value_websocket(): return wrapTheFunction +def algorithm_kill_value_websocket(): + """ + 获取kill值, websocket发布 + """ + + def wrapTheFunction(func): + @wraps(func) + def wrapped_function(*args, **kwargs): + data = func(*args, **kwargs) + id = data["id"] + data_res = {'code': 1, "type": 'kill', 'msg': 'success', 'data': data} + manager.send_message_proj_json(message=data_res, id=id) + return data + + return wrapped_function + + return wrapTheFunction + + +def algorithm_error_value_websocket(): + """ + 获取error值, websocket发布 + """ + + def wrapTheFunction(func): + @wraps(func) + def wrapped_function(*args, **kwargs): + data = func(*args, **kwargs) + id = data["id"] + data_res = {'code': 2, "type": 'error', 'msg': 'fail', 'data': data} + manager.send_message_proj_json(message=data_res, id=id) + return data + + return wrapped_function + + return wrapTheFunction def obtain_train_param(): """ @@ -164,7 +201,6 @@ def obtain_train_param(): return wrapTheFunction - def obtain_test_param(): """ 获取验证参数 @@ -215,6 +251,16 @@ def obtain_download_pt_param(): return wrapTheFunction +@bp.route('/change_ifKillDIct', methods=['get']) +def change_ifKillDIct(): + """ + 修改全局变量 + """ + id = request.args.get('id') + type = request.args.get('type') + global ifKillDict + ifKillDict[id] = False + return output_wrapped(0, 'success') # @start_train_algorithm() # def start(param: str): @@ -241,6 +287,13 @@ from app.schemas.TrainResult import DetectProcessValueDice, DetectReport from app import file_tool +def error_return(id: str, data): + """ + 算法出错，返回 + """ + data_res = {'code': 2, "type": 'error', 'msg': 'fail', 'data': data} + manager.send_message_proj_json(message=data_res, id=id) + # 启动训练 @start_train_algorithm() def train_R0DY(params_str, id): @@ -255,8 +308,12 @@ def train_R0DY(params_str, id): epoches = params.get('epochnum').value batch_size = params.get('batch_size').value device = params.get('device').value - - train_start(weights, savemodel, epoches, img_size, batch_size, device, data_list, id) + try: + train_start(weights, savemodel, epoches, img_size, batch_size, device, data_list, id) + print("train down!") + except Exception as e: + print(repr(e)) + error_return(id=id,data=repr(e)) # 启动验证程序 @@ -303,7 +360,8 @@ def Export_model_RODY(params_str): exp_inputPath = params.get('exp_inputPath').value # 模型路径 print('输入模型：', exp_inputPath) exp_device = params.get('device').value - modellist = Start_Model_Export(exp_inputPath, exp_device) + imgsz = params.get('imgsz').value + modellist = Start_Model_Export(exp_inputPath, exp_device, imgsz) exp_outputPath = exp_inputPath.replace('pt', 'zip') # 压缩文件 print('模型路径：',exp_outputPath) zipf = zipfile.ZipFile(exp_outputPath, 'w') @@ -312,20 +370,19 @@ def Export_model_RODY(params_str): return exp_outputPath - @obtain_train_param() def returnTrainParams(): - # nvmlInit() - # gpuDeviceCount = nvmlDeviceGetCount() # 获取Nvidia GPU块数 - # _kernel = [f"cuda:{a}" for a in range(gpuDeviceCount)] + nvmlInit() + gpuDeviceCount = nvmlDeviceGetCount() # 获取Nvidia GPU块数 + _kernel = [f"cuda:{a}" for a in range(gpuDeviceCount)] params_list = [ {"index": 0, "name": "epochnum", "value": 10, "description": '训练轮次', "default": 100, "type": "I", 'show': True}, {"index": 1, "name": "batch_size", "value": 4, "description": '批次图像数量', "default": 1, "type": "I", 'show': True}, {"index": 2, "name": "img_size", "value": 640, "description": '训练图像大小', "default": 640, "type": "I", 'show': True}, - {"index": 3, "name": "device", "value": "0", "description": '训练核心', "default": "cuda", "type": "S", - "items": '', 'show': True}, # _kernel + {"index": 3, "name": "device", "value": 'CUDA', "description": '训练核心', "default": 'CUDA', "type": "E", + "items": _kernel, 'show': True}, # _kernel {"index": 4, "name": "saveModDir", "value": "E:/alg_demo-master/alg_demo/app/yolov5/best.pt", "description": '保存模型路径', "default": "./app/maskrcnn/saved_model/test.pt", "type": "S", 'show': False}, @@ -381,7 +438,7 @@ def returnDetectParams(): {"index": 1, "name": "outputPath", "value": 'E:/aicheck/data_set/11442136178662604800/val_results/', "description": '输出结果路径', "default": './app/maskrcnn/datasets/M006B_waibi/res', "type": "S", 'show': False}, - {"index": 0, "name": "modPath", "value": "E:/alg_demo-master/alg_demo/app/yolov5/圆孔_123_RODY_1_640.pt", + {"index": 2, "name": "modPath", "value": "E:/alg_demo-master/alg_demo/app/yolov5/圆孔_123_RODY_1_640.pt", "description": '模型路径', "default": "./app/maskrcnn/saved_model/test.pt", "type": "S", 'show': False}, {"index": 3, "name": "device", "value": "0", "description": '推理核', "default": "cpu", "type": "S", 'show': False}, @@ -399,7 +456,9 @@ def returnDownloadParams(): "default": 'E:/alg_demo-master/alg_demo/app/yolov5/圆孔_123_RODY_1_640.pt/', "type": "S", 'show': False}, {"index": 1, "name": "device", "value": 'gpu', "description": 'CPU或GPU', "default": 'gpu', "type": "S", - 'show': False} + 'show': False}, + {"index": 2, "name": "imgsz", "value": 640, "description": '图像大小', "default": 640, "type": "I", + 'show': True} ] params_str = json.dumps(params_list) return params_str diff --git a/app/schemas/TrainResult.py b/app/schemas/TrainResult.py index a5bb7a0..ff07135 100644 --- a/app/schemas/TrainResult.py +++ b/app/schemas/TrainResult.py @@ -32,6 +32,7 @@ class Report(BaseModel): train_mod_savepath: str = Field(..., description="模型保存路径") start_time: datetime.date = Field(datetime.datetime.now(), description="开始时间") end_time: datetime.date = Field(datetime.datetime.now(), description="结束时间") + alg_code: str = Field(..., description="模型编码") class ReportDict(BaseModel): diff --git a/app/yolov5/export.py b/app/yolov5/export.py index ce12ab7..89ba977 100644 --- a/app/yolov5/export.py +++ b/app/yolov5/export.py @@ -566,11 +566,12 @@ def run( return f # return list of exported files/dirs -def parse_opt(weights,device): +def parse_opt(weights,device,imgsz): + imgsz = [imgsz,imgsz] parser = argparse.ArgumentParser() parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='dataset.yaml path') parser.add_argument('--weights', nargs='+', type=str, default=weights, help='model.pt path(s)') - parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640, 640], help='image (h, w)') + parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=imgsz, help='image (h, w)') #default=[640, 640] parser.add_argument('--batch-size', type=int, default=1, help='batch size') parser.add_argument('--device', default=device, help='cuda device, i.e. 0 or 0,1,2,3 or cpu') parser.add_argument('--half', action='store_true', help='FP16 half-precision export') @@ -604,13 +605,13 @@ def main(opt): f = run(**vars(opt)) return f -def Start_Model_Export(weights,device): +def Start_Model_Export(weights,device,imgsz): # 判断cpu or gpu if device == 'gpu': device = '0' else: device = 'cpu' - opt = parse_opt(weights,device) + opt = parse_opt(weights,device,imgsz) f = main(opt) return f diff --git a/app/yolov5/train_server.py b/app/yolov5/train_server.py index 07ff125..9a625a7 100644 --- a/app/yolov5/train_server.py +++ b/app/yolov5/train_server.py @@ -61,6 +61,8 @@ from app.yolov5.utils.torch_utils import (EarlyStopping, ModelEMA, de_parallel, smart_resume, torch_distributed_zero_first) from app.schemas.TrainResult import Report, ProcessValueList from app.controller.AlgorithmController import algorithm_process_value_websocket +from app.configs import global_var +from app.utils.websocket_tool import manager LOCAL_RANK = int(os.getenv('LOCAL_RANK', -1)) # https://pytorch.org/docs/stable/elastic/run.html RANK = int(os.getenv('RANK', -1)) WORLD_SIZE = int(os.getenv('WORLD_SIZE', 1)) @@ -72,7 +74,7 @@ def yaml_rewrite(file='data.yaml',data_list=[]): with open(file, errors='ignore') as f: coco_dict = yaml.safe_load(f) #读取img_label_type.json - with open(data_list[3], 'r') as f: + with open(data_list[3], 'r',encoding='UTF-8') as f: class_dict = json.load(f) f.close() classes = class_dict["classes"] @@ -302,7 +304,17 @@ def train(hyp, opt, device, data_list,id,callbacks): # hyp is path/to/hyp.yaml report = Report(rate_of_progess=0, precision=[process_value_list], id=id, sum=epochs, progress=0, num_train_img=train_num, - train_mod_savepath=best) + train_mod_savepath=best, + alg_code="R-ODY") + + def kill_return(): + """ + 算法中断，返回 + """ + id = report.id + data = report.dict() + data_res = {'code': 1, "type": 'kill', 'msg': 'fail', 'data': data} + manager.send_message_proj_json(message=data_res, id=id) @algorithm_process_value_websocket() def report_cellback(i, num_epochs, reportAccu): @@ -314,6 +326,12 @@ def train(hyp, opt, device, data_list,id,callbacks): # hyp is path/to/hyp.yaml ###################结束####################### for epoch in range(start_epoch, epochs): # epoch ------------------------------------------------------------------ #callbacks.run('on_train_epoch_start') + print("start get global_var") + ifkill = global_var.get_value(report.id) + print("get global_var down:",ifkill) + if ifkill: + kill_return() + break model.train() # Update image weights (optional, single-GPU only) @@ -334,6 +352,12 @@ def train(hyp, opt, device, data_list,id,callbacks): # hyp is path/to/hyp.yaml optimizer.zero_grad() for i, (imgs, targets, paths, _) in pbar: # batch ------------------------------------------------------------- #callbacks.run('on_train_batch_start') + print("start get global_var") + ifkill = global_var.get_value(report.id) + print("get global_var down:",ifkill) + if ifkill: + kill_return() + break if targets.shape[0] == 0: targets = [[0.00000, 5.00000, 0.97002, 0.24679, 0.05995, 0.05553], [0.00000, 7.00000, 0.95097, 0.32007, 0.04188, 0.02549], diff --git a/app/yolov5/utils/dataloaders.py b/app/yolov5/utils/dataloaders.py index 90de966..9b8a9b8 100644 --- a/app/yolov5/utils/dataloaders.py +++ b/app/yolov5/utils/dataloaders.py @@ -430,9 +430,12 @@ class LoadImagesAndLabels(Dataset): self.label_files = img2label_paths(self.im_files) # labels cache_path = (p if p.is_file() else Path(self.label_files[0]).parent).with_suffix('.cache') try: - cache, exists = np.load(cache_path, allow_pickle=True).item(), True # load dict - assert cache['version'] == self.cache_version # matches current version - assert cache['hash'] == get_hash(self.label_files + self.im_files) # identical has + # cache, exists = np.load(cache_path, allow_pickle=True).item(), True # load dict + # assert cache['version'] == self.cache_version # matches current version + # assert cache['hash'] == get_hash(self.label_files + self.im_files) # identical has + if os.path.exists(cache_path): + os.remove(cache_path) + cache, exists = self.cache_labels(cache_path, prefix), False # run cache ops except Exception: cache, exists = self.cache_labels(cache_path, prefix), False # run cache ops