We have trouble with debugging SC3 breast main lane. I wonder SC3 breast proteome matrix are different shape(N of proteins, retro&ex=10006 but main=10005??). And I want to know this error log. I'm sorry to bother you over and over. -- Hello Yasuhiro Kambara, Sorry, but we were unable to process your submission to the Proteogenomics Subchallenge 3 Breast. Please refer to the challenge instructions which can be found at https://www.synapse.org/#!Synapse:syn8228304/wiki/448381 and to the error message below: submission name: syn11456497 submission ID: 9654108

Created by Yasuhiro Kambara kambarakun
Please see: https://www.synapse.org/#!Synapse:syn8228304/discussion/threadId=2922
Thanks for reply! I had understood sample's names are different, but are the numbers of proteins different too? I wonder "prospective_breast_proteome_sort_common_gene_10005.txt" is shorten or don't include header or something. Anyways, *_breast_proteome_sort_common_gene_"10005" includes 10006 proteins on retro and ex!
Yes the dimensions of the matrix are different form the express lanes data, the sample names are different, and the main data will have NA's. Hopefully taking these into account will help you debug. Your error log: /sc3_model_and_functions.py:21: RuntimeWarning: invalid value encountered in subtract x_numer = x - x_base.min(axis=0) /sc3_model_and_functions.py:22: RuntimeWarning: invalid value encountered in subtract x_denom = xp.max(xp.hstack([(x_base.max(axis=0) - x_base.min(axis=0)), xp.array(xp.repeat(xp.finfo(float).eps, repeats=x_base.shape[1]).astype(xp.float32))]), axis=0) Traceback (most recent call last): File "/sc3_predict.py", line 100, in list_predict.append(model_pro_to_pho.predict(xp_pro)) File "/sc3_model_and_functions.py", line 394, in predict return self.chain_common_to_pho(self.chain_pro_to_common(x)) File "/sc3_model_and_functions.py", line 104, in __call__ self.h_0 = self.bn_input(x) File "/usr/local/lib/python3.5/dist-packages/chainer/links/normalization/batch_normalization.py", line 143, in __call__ running_var=self.avg_var, decay=decay) File "/usr/local/lib/python3.5/dist-packages/chainer/functions/normalization/batch_normalization.py", line 525, in batch_normalization (x, gamma, beta))[0] File "/usr/local/lib/python3.5/dist-packages/chainer/function_node.py", line 206, in apply self._check_data_type_forward(in_data) File "/usr/local/lib/python3.5/dist-packages/chainer/function_node.py", line 274, in _check_data_type_forward self.check_type_forward(in_type) File "/usr/local/lib/python3.5/dist-packages/chainer/functions/normalization/batch_normalization.py", line 46, in check_type_forward gamma_type.shape == beta_type.shape, File "/usr/local/lib/python3.5/dist-packages/chainer/utils/type_check.py", line 514, in expect expr.expect() File "/usr/local/lib/python3.5/dist-packages/chainer/utils/type_check.py", line 472, in expect '{0} {1} {2}'.format(left, self.inv, right)) chainer.utils.type_check.InvalidType: Invalid operation is performed in: BatchNormalization (Forward) Expect: in_types[0].shape[1:2] == in_types[1].shape Actual: (10005,) != (10006,)

Question of SC3 main lane's matrix(rows x columns) page is loading…