GoogLeNet Ignoring source layer

[data_garden]

I'm running a Caffe framework (CPU) inside a docker container, with the following piece of code:

 

    model_path = 'models/bvlc_googlenet/' # substitute your path here
    net_fn   = model_path + 'deploy.prototxt'
    param_fn = model_path + 'bvlc_googlenet.caffemodel'
   
    # Patching model to be able to compute gradients.
    # Note that you can also manually add "force_backward: true" line to "deploy.prototxt".
    model = caffe.io.caffe_pb2.NetParameter()
    text_format.Merge(open(net_fn).read(), model)
    model.force_backward = True
    open('models/bvlc_googlenet/tmp.prototxt', 'w').write(str(model))
   
   
    net = caffe.Classifier('tmp.prototxt', param_fn,
                           mean = np.float32([104.0, 116.0, 122.0]), # ImageNet mean, training set dependent
                           channel_swap = (2,1,0)) # the reference model has channels in BGR order instead of RGB

but, after having all model layers ignored, I' getting the following traceback:

   

 libdc1394 error: Failed to initialize libdc1394
    WARNING: Logging before InitGoogleLogging() is written to STDERR
    I0227 00:08:49.033560   122 net.cpp:49] Initializing net from parameters:
    force_backward: true
    state {
      phase: TEST
    }
    I0227 00:08:49.033826   122 net.cpp:274] Network initialization done.
    I0227 00:08:49.316442   122 upgrade_proto.cpp:52] Attempting to upgrade input file specified using deprecated V1LayerParameter: models/bvlc_googlenet/bvlc_googlenet.caffemodel
    I0227 00:08:49.370679   122 upgrade_proto.cpp:60] Successfully upgraded file specified using deprecated V1LayerParameter
    I0227 00:08:49.374241   122 net.cpp:752] Ignoring source layer data
    I0227 00:08:49.374426   122 net.cpp:752] Ignoring source layer label_data_1_split
    I0227 00:08:49.374480   122 net.cpp:752] Ignoring source layer conv1/7x7_s2
    I0227 00:08:49.374514   122 net.cpp:752] Ignoring source layer conv1/relu_7x7
    I0227 00:08:49.374563   122 net.cpp:752] Ignoring source layer pool1/3x3_s2
    I0227 00:08:49.374616   122 net.cpp:752] Ignoring source layer pool1/norm1
    I0227 00:08:49.374683   122 net.cpp:752] Ignoring source layer conv2/3x3_reduce
    (for all layers)
    Traceback (most recent call last):
      File "dream.py", line 40, in <module>
        channel_swap = (2,1,0)) # the reference model has channels in BGR order instead of RGB
      File "/root/caffe/python/caffe/classifier.py", line 29, in __init__
        in_ = self.inputs[0]
    IndexError: list index out of range

could anyone please point me in the right direction?