pre trained model of CUB?

FzuLJ commented 6 years ago

Can you please provide a pre trained model for CUB-200-2011?

xiaomengyc commented 5 years ago

@FzuLJ
Sorry for the late reply. The trained model on CUB dataset has been uploaded at https://drive.google.com/open?id=1WfrELBlEoq5WO7gKUv-MLTQ8QHY-2wiX

Rahul-Venugopal commented 5 years ago

When I run sh val_cub_full.sh , I am getting following output / Error

Project Root Dir: /home/rahul/Desktop/Thesis/Reference/WSOL/ACoL/SPG Running parameters:

{ "root_dir":"/home/rahul/Desktop/Thesis/Reference/WSOL/ACoL/SPG", "train_list":"/home/rahul/Desktop/Thesis/Reference/WSOL/ACoL/SPG/datalist/CUB/train_list.txt", "tencrop":"False", "threshold":0.5, "onehot":"True", "restore_from":"", "test_list":"/home/rahul/Desktop/Thesis/Reference/WSOL/ACoL/SPG/datalist/CUB/test_list.txt", "num_gpu":1, "crop_size":321, "disp_interval":20, "resume":"True", "snapshot_dir":"../snapshots/inception_imagenet_full_spg/", "epoch":21, "lr":0.001, "input_size":356, "current_epoch":0, "num_workers":20, "batch_size":1, "global_counter":0, "dataset":"cub", "arch":"inception3_spg", "num_classes":200, "img_dir":"/home/rahul/Desktop/Thesis/Reference/WSOL/ACoL/SPG/data/CUB_200_2011/images", "decay_points":"none" }

../models/google/inception3_spg.py:137: UserWarning: nn.init.xavier_uniform is now deprecated in favor of nn.init.xavier_uniform_.
  nn.init.xavier_uniform(m.weight.data)
=> no checkpoint found at ''
0it [00:00, ?it/s]/home/rahul/.virtualenvs/acol/local/lib/python2.7/site-packages/torch/nn/modules/upsampling.py:122: UserWarning: nn.Upsampling is deprecated. Use nn.functional.interpolate instead.
  warnings.warn("nn.Upsampling is deprecated. Use nn.functional.interpolate instead.")
/home/rahul/.virtualenvs/acol/local/lib/python2.7/site-packages/torch/nn/functional.py:1961: UserWarning: Default upsampling behavior when mode=bilinear is changed to align_corners=False since 0.4.0. Please specify align_corners=True if the old behavior is desired. See the documentation of nn.Upsample for details.
  "See the documentation of nn.Upsample for details.".format(mode))

Traceback (most recent call last):
  File "val_frame.py", line 203, in <module>
    val(args)
  File "val_frame.py", line 132, in val
    val_mAP, prob, gt = cal_mAP(logits0, label_var, prob, gt)
  File "val_frame.py", line 178, in cal_mAP
    assert logits0.size() == label_var.size()
AssertionError

Any suggestions about the mistake I am making ?

xiaomengyc commented 5 years ago

@Rahul-Venugopal You got to change to onehot=False if your images do not have multiclasses.

xiaomengyc / SPG

pre trained model of CUB? #4