I assume you mean how to define phase when writing a prototxt using caffe.NetSpec? from caffe import layers as L, params as P, to_proto import caffe ns = caffe.NetSpec() ns.data = L.Data(name=”data”, data_param={‘source’:’/path/to/lmdb’,’batch_size’:32}, include={‘phase’:caffe.TEST}) If you want to have BOTH train and test layers in the same prototxt, what I usually do is making one … Read more
It is a common practice to decrease the learning rate (lr) as the optimization/learning process progresses. However, it is not clear how exactly the learning rate should be decreased as a function of the iteration number. If you use DIGITS as an interface to Caffe, you will be able to visually see how the different … Read more
Some linear-algebra background Singular Value Decomposition (SVD) is a decomposition of any matrix W into three matrices: W = U S V* Where U and V are ortho-normal matrices, and S is diagonal with elements in decreasing magnitude on the diagonal. One of the interesting properties of SVD is that it allows to easily approximate … Read more
There are two main differences between a “train” prototxt and a “deploy” one: 1. Inputs: While for training data is fixed to a pre-processed training dataset (lmdb/HDF5 etc.), deploying the net require it to process other inputs in a more “random” fashion. Therefore, the first change is to remove the input layers (layers that push … Read more
You are correct in observing that the scaling you are doing for the log(prob) is exactly what “InfogainLoss” layer is doing (You can read more about it here and here). As for the derivative (back-prop): the loss computed by this layer is L = – sum_j infogain_mat[label * dim + j] * log( prob(j) ) … Read more
You can change the iter_size in the solver parameters. Caffe accumulates gradients over iter_size x batch_size instances in each stochastic gradient descent step. So increasing iter_size can also get more stable gradient when you cannot use large batch_size due to the limited memory.
Assuming you are trying to do image classification. These should be the steps for finetuning a model: 1. Classification layer The original classification layer “loss3/classifier” outputs predictions for 1000 classes (it’s mum_output is set to 1000). You’ll need to replace it with a new layer with appropriate num_output. Replacing the classification layer: Change layer’s name … Read more
Why don’t you use the InfogainLoss layer to compensate for the imbalance in your training set? The Infogain loss is defined using a weight matrix H (in your case 2-by-2) The meaning of its entries are [cost of predicting 1 when gt is 0, cost of predicting 0 when gt is 0 cost of predicting … Read more
Should you shuffle the samples? Think about the learning process if you don’t shuffle; caffe sees only 0 samples – what do you expect the algorithm to deduce? simply predict 0 all the time and everything is cool. If you have plenty of 0 before you hit the first 1 caffe will be very confident … Read more
You can use a “Python” layer: a layer implemented in python to feed data into your net. (See an example for adding a type: “Python” layer here). import sys, os sys.path.insert(0, os.environ[‘CAFFE_ROOT’]+’/python’) import caffe class myInputLayer(caffe.Layer): def setup(self,bottom,top): # read parameters from `self.param_str` … def reshape(self,bottom,top): # no “bottom”s for input layer if len(bottom)>0: raise … Read more