I had an AutoAnnotation script for yolo v7, which is part of a annotation tool. When I call an api, it take the dataset, weights path and all required details, and publish it via rabbitmq to a gpu server, In gpu server, there is a call back instance, which is loading the model to do auto annotation.
I am loading the model using YOLOv7 –>models–>experimental–>attempt_load function, where Iam successfully loading the function, but

def attempt_load(weights, map_location=None):
    # Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a
    model = Ensemble()
    for w in weights if isinstance(weights, list) else [weights]:
        attempt_download(w)
        ckpt = torch.load(w, map_location=map_location)  # load
        model.append(ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval())  # FP32 model
    
    # Compatibility updates
    for m in model.modules():
        if type(m) in [nn.Hardswish, nn.LeakyReLU, nn.ReLU, nn.ReLU6, nn.SiLU]:
            m.inplace = True  # pytorch 1.7.0 compatibility
        elif type(m) is nn.Upsample:
            m.recompute_scale_factor = None  # torch 1.11.0 compatibility
        elif type(m) is Conv:
            m._non_persistent_buffers_set = set()  # pytorch 1.6.0 compatibility
    
    if len(model) == 1:
        return model[-1]  # return model
    else:
        print('Ensemble created with %sn' % weights)
        for k in ['names', 'stride']:
            setattr(model, k, getattr(model[-1], k))
        return model  # return ensemble

the python instance is always breaking at the below line

model.append(ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval())  # FP32 model

but if, I run the same instance in the python console manually, It’s working fine and completing auto annotation. I kept logs but , There are no any exceptions found. What could be the reason. I didn’t understand this behaviour.

This same issue is occuring with different gpu servers.

//

This issue is new to me, I never faced any thing like this, I tried for searching, similar kind of issue over internet, but I didn’t find any solution.

Thanks for any advice in advance.