dicarlolab / archconvnets

Architecturally optimized neural networks trained with regularized backpropagation
5 stars 5 forks source link

hdf5 read mode #28

Open ardila opened 10 years ago

ardila commented 10 years ago

@yamins81 I believe this error is due to not having the correct skdata version. Where is the correct version?

Loading checkpoint from database.
Loading checkpoint from "recent" storage.
Can't import separate mcc package
Can't import asgd.
Feature database not configured
cannot connect to db on port  27017
Getting stimuli from cache hdf5 at /om/user/yamins/.skdata/imagenet/ChallengeSynsets2013_offline_23d6ee636ade8ad3912204410d1acc23c10357cf/cache/images_cache_136f0d237e71c46fa47a71becf1f5d0fd5693ce6_random_0_hdf5
Traceback (most recent call last):
  File "convnet.py", line 320, in <module>
    model = ConvNet(op, load_dic)
  File "convnet.py", line 119, in __init__
    IGPUModel.__init__(self, "ConvNet", op, load_dic, filename_options, dp_params=dp_params)
  File "/home/ardila/src/archconvnets/archconvnets/convnet2/python_util/gpumodel.py", line 186, in __init__
    self.init_data_providers()
  File "/home/ardila/src/archconvnets/archconvnets/convnet2/python_util/gpumodel.py", line 214, in init_data_providers
    type=self.dp_type, dp_params=self.dp_params, test=True)
  File "/home/ardila/src/archconvnets/archconvnets/convnet2/python_util/data.py", line 124, in get_instance
    return _class(data_dir, batch_range, init_epoch, init_batchnum, dp_params, test)
  File "/home/ardila/src/archconvnets/archconvnets/convnet2/convdata.py", line 321, in __init__
    DLDataProvider2.__init__(self, data_dir, batch_range, init_epoch, init_batchnum, dp_params, test, cache_type='hdf5', read_mode='r')
  File "/home/ardila/src/archconvnets/archconvnets/convnet2/python_util/data.py", line 553, in __init__
    mode=read_mode)
TypeError: __init__() got an unexpected keyword argument 'mode'