(torchLFS) [inspur@localhost LibFewShot-main]$ python run_trainer.py [11/14/22 10:04:36] INFO {'data_root': '/home/inspur/MAX_SPACE/Jiyu/LibFewShot-main/dataset_floder/miniImgFSL', 'image_size': 84, 'use_memory': False, 'augment': trainer.py:372 False, 'augment_times': 1, 'augment_times_query': 1, 'workers': 2, 'dataloader_num': 1, 'device_ids': 0, 'n_gpu': 1, 'seed': 2147483647,
'deterministic': True, 'port': 53758, 'log_name': None, 'log_level': 'info', 'log_interval': 100, 'log_paramerter': False, 'result_root':
'./results', 'save_interval': 10, 'save_part': ['emb_func'], 'tag': None, 'epoch': 5, 'test_epoch': 5, 'parallel_part': ['emb_func'],
'pretrain_path': None, 'resume': False, 'way_num': 5, 'shot_num': 1, 'query_num': 15, 'test_way': 5, 'test_shot': 1, 'test_query': 15,
'episode_size': 1, 'train_episode': 100, 'test_episode': 100, 'batch_size': 128, 'val_per_epoch': 1, 'optimizer': {'name': 'Adam',
'kwargs': {'lr': 0.01}, 'other': None}, 'lr_scheduler': {'name': 'StepLR', 'kwargs': {'gamma': 1.0, 'step_size': 20}}, 'warmup': 0,
'includes': ['headers/data.yaml', 'headers/device.yaml', 'headers/misc.yaml', 'headers/model.yaml', 'headers/optimizer.yaml',
'classifiers/RENet.yaml', 'backbones/resnet18.yaml'], 'classifier': {'name': 'RENet', 'kwargs': {'feat_dim': 512, 'lambda_epi': 0.25,
'temperature': 0.2, 'temperature_attn': 5.0, 'num_classes': 2}}, 'backbone': {'name': 'resnet18', 'kwargs': {'is_feature': False,
'avg_pool': True, 'is_flatten': True}}, 'tb_scale': 1.0, 'rank': 0}
[11/14/22 10:04:37] INFO RENet( trainer.py:372 (emb_func): ResNet(
(conv1): Conv2d(3, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(layer1): Sequential(
(0): BasicBlock(
(conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(1): BasicBlock(
(conv1): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(layer2): Sequential(
(0): BasicBlock(
(conv1): Conv2d(64, 128, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(downsample): Sequential(
(0): Conv2d(64, 128, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): BasicBlock(
(conv1): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(128, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(128, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(layer3): Sequential(
(0): BasicBlock(
(conv1): Conv2d(128, 256, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(downsample): Sequential(
(0): Conv2d(128, 256, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): BasicBlock(
(conv1): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(256, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(layer4): Sequential(
(0): BasicBlock(
(conv1): Conv2d(256, 512, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(downsample): Sequential(
(0): Conv2d(256, 512, kernel_size=(1, 1), stride=(2, 2), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(1): BasicBlock(
(conv1): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(conv2): Conv2d(512, 512, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
(avgpool): AdaptiveAvgPool2d(output_size=(1, 1))
)
(fc): Linear(in_features=512, out_features=2, bias=True)
(scr_layer): SCRLayer(
(model): Sequential(
(0): SelfCorrelationComputation(
(unfold): Unfold(kernel_size=(5, 5), dilation=1, padding=2, stride=1)
(relu): ReLU()
)
(1): SCR(
(conv1x1_in): Sequential(
(0): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(conv1): Sequential(
(0): Conv3d(64, 64, kernel_size=(1, 3, 3), stride=(1, 1, 1), bias=False)
(1): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(conv2): Sequential(
(0): Conv3d(64, 64, kernel_size=(1, 3, 3), stride=(1, 1, 1), bias=False)
(1): BatchNorm3d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU(inplace=True)
)
(conv1x1_out): Sequential(
(0): Conv2d(64, 512, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(512, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
)
)
)
(cca_layer): CCALayer(
(cca_module): CCA(
(conv): Sequential(
(0): SepConv4d(
(proj): Sequential(
(0): Conv2d(1, 16, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(conv1): Sequential(
(0): Conv3d(1, 1, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1), bias=False)
(1): BatchNorm3d(1, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(conv2): Sequential(
(0): Conv3d(1, 1, kernel_size=(3, 3, 1), stride=(1, 1, 1), padding=(1, 1, 0), bias=False)
(1): BatchNorm3d(1, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(relu): ReLU(inplace=True)
)
(1): ReLU(inplace=True)
(2): SepConv4d(
(proj): Sequential(
(0): Conv2d(16, 1, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(1, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(conv1): Sequential(
(0): Conv3d(16, 16, kernel_size=(1, 3, 3), stride=(1, 1, 1), padding=(0, 1, 1), bias=False)
(1): BatchNorm3d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(conv2): Sequential(
(0): Conv3d(16, 16, kernel_size=(3, 3, 1), stride=(1, 1, 1), padding=(1, 1, 0), bias=False)
(1): BatchNorm3d(16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
)
(relu): ReLU(inplace=True)
)
)
)
(cca_1x1): Sequential(
(0): Conv2d(512, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
(1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(2): ReLU()
)
)
(loss_func): CrossEntropyLoss()
)
INFO Trainable params in the model: 11348186 trainer.py:372 [11/14/22 10:04:40] INFO load 38400 train image with 64 label. trainer.py:372 INFO load 9600 val image with 16 label. trainer.py:372 INFO load 12000 test image with 20 label. trainer.py:372 INFO Adam ( trainer.py:372 Parameter Group 0
amsgrad: False
betas: (0.9, 0.999)
eps: 1e-08
initial_lr: 0.01
lr: 0.01
weight_decay: 0
)
INFO ============ Train on the train set ============ trainer.py:372 INFO learning rate: [0.01] trainer.py:372 Traceback (most recent call last): File "run_trainer.py", line 24, in main(0, config) File "run_trainer.py", line 14, in main trainer.train_loop(rank) File "/home/inspur/MAX_SPACE/Jiyu/LibFewShot-main/core/trainer.py", line 83, in train_loop train_acc = self._train(epoch_idx) File "/home/inspur/MAX_SPACE/Jiyu/LibFewShot-main/core/trainer.py", line 170, in _train [elem for each_batch in batch for elem in each_batch] File "/home/inspur/anaconda3/envs/torchLFS/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl result = self.forward(*input, **kwargs) File "/home/inspur/MAX_SPACE/Jiyu/LibFewShot-main/core/model/abstract_model.py", line 30, in forward return self.set_forward_loss(x) File "/home/inspur/MAX_SPACE/Jiyu/LibFewShot-main/core/model/finetuning/renet.py", line 415, in set_forward_loss ) = batch # RENet uses both episode and general dataloaders ValueError: not enough values to unpack (expected 4, got 2)

Hi. I have installed LibFewShot successfully, and got correct output by running test_install.yaml. However, the script gave the following error message when I tried to run other supported methods listed in ./config. Could you give some suggestions about how to solve such problem?

Hello, as described in the title, how do I handle my own dataset after training with the miniImagenet dataset you have given ? For example,my dataset has 4 classes and each class has about 100 images, Do I need to process my dataset into the following format?

MyDataset_folder/ ├── images/ │ ├── images_1.jpg │ ├── ... │ └── images_n.jpg ├── train.csv * ├── test.csv * └── val.csv *

Or do I need to put all my datasets into the miniImagenet dataset and create a new test.csv to replace the original test.csv in the miniImagenet dataset?

Or other methods?

Look forward to your reply, thank you!

Hi,

I'm trying to use your library with a custom dataset that I created with images retreived from google image. The train/val.csv file is composed of seven classes each with 5 examples and the test.csv is composed of one image of each classes (see below pictures).

I have also the following .yaml config file that I -more or less - copy-pasted from your documentation,

includes:
  - headers/dataZSL.yaml
  - headers/device.yaml
  - headers/misc.yaml
  - headers/model.yaml
  - headers/optimizer.yaml
  - classifiers/Proto.yaml
  - backbones/Conv64F.yaml


epoch: 50
test_epoch: 5

pretrain_path: ~
resume: False

way_num: 5
shot_num: 5
query_num: 15
test_way: ~
test_shot: ~
test_query: ~
episode_size: 1
# batch_size only works in pre-train
batch_size: 128
train_episode: 10000
test_episode: 1000

backbone:
  name: Conv64F
  kwargs:
    is_flatten: True
    is_feature: False
    leaky_relu: False
    negative_slope: 0.2
    last_pool: True

the dataZSL.yaml file referenced in the header is the following file,

data_root: ./core/data/fewshot/zslDataset
image_size: 84
use_memory: False
augment: True
augment_times: 1
augment_times_query: 1

If that is useful, I will precise that I'm using google colab to run the code. With all that said, when I try to lunch the run_trainer.py program the following error occurs,

Traceback (most recent call last): File "run_trainer.py", line 12, in trainer.train_loop() File "/content/drive/MyDrive/Colab Notebooks/LibFewShot/core/trainer.py", line 72, in train_loop train_acc = self._train(epoch_idx) File "/content/drive/MyDrive/Colab Notebooks/LibFewShot/core/trainer.py", line 119, in _train prefetcher = data_prefetcher(self.train_loader) File "/content/drive/MyDrive/Colab Notebooks/LibFewShot/core/utils/utils.py", line 288, in init self.preload() File "/content/drive/MyDrive/Colab Notebooks/LibFewShot/core/utils/utils.py", line 292, in preload self.next_data = next(self.loader) File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py", line 530, in next data = self._next_data() File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py", line 1224, in _next_data return self._process_data(data) File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/dataloader.py", line 1250, in _process_data data.reraise() File "/usr/local/lib/python3.7/dist-packages/torch/_utils.py", line 457, in reraise raise exception RuntimeError: Caught RuntimeError in DataLoader worker process 0. Original Traceback (most recent call last): File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/worker.py", line 287, in _worker_loop data = fetcher.fetch(index) File "/usr/local/lib/python3.7/dist-packages/torch/utils/data/_utils/fetch.py", line 52, in fetch return self.collate_fn(data) File "/content/drive/MyDrive/Colab Notebooks/LibFewShot/core/data/collates/collate_functions.py", line 169, in call return self.method(batch) File "/content/drive/MyDrive/Colab Notebooks/LibFewShot/core/data/collates/collate_functions.py", line 149, in method self.episode_size, self.way_num, self.shot_num + self.query_num RuntimeError: shape '[1, 5, 20]' is invalid for input of size 25

I must say that I am quite lost in how to resolve this problem. If I can provide more information, don't hesitate to ask, Thank you.

the two first classes in train.csv

the test.csv file

PS : I removed the first column in the lib code in dataset.py in order to only get the two columns of interest

Hi, I tried to reproduce the accuracy of CAN method. However, the accuracy turns out to be weird. So that I print the intermediate steps and the returned value once line 327 in can.py is called.

acc = accuracy(cls_scores, query_targets, topk=1)

One example of output:

batch_size:  1
correct:  tensor([[1, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0, 0, 1, 0, 0, 1, 1, 1, 0, 1, 0, 0, 1,
         1, 1, 0, 1, 0, 1, 1, 1, 0, 1, 0, 1, 1, 0, 1, 0, 1, 0, 0, 0, 0, 1, 1, 0,
         0, 1, 1, 0, 1, 1, 0, 0, 0, 0, 1, 1, 0, 0, 0, 1, 0, 0, 1, 1, 0, 0, 0, 0,
         0, 0, 0]], device='cuda:0', dtype=torch.uint8)
correct_k:  tensor([30.], device='cuda:0')
 acc:  3000.0

The accuracy is too large. Also, the batch_size is set as 128 in the config file, rather than the output value 1.

你好，首先感谢你们对多种小样本方法的复现及开源。

在运行leo.yaml配置时我注意到它需要pretrain的模型：pretrain_path: ./results/checkpoint.pth。请问这个模型是用什么方法预训练的呢？

另外在运行其它方法时，根据给出的配置我跑出了和论文差距不小的结果，我注意到您在 #9 中提到过将放出具体配置，请问目前有哪些方法已经整理好了配置，能否更新一下，十分感谢！

maml方法能在单个gpu上训练，但在多个gpu上平行训练会报错。具体错误如下：

  File "/home/cougarnet.uh.edu/pyuan2/anaconda3/envs/py37/lib/python3.7/site-packages/torch/nn/parallel/parallel_apply.py", line 61, in _worker
    output = module(*input, **kwargs)
  File "/home/cougarnet.uh.edu/pyuan2/anaconda3/envs/py37/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/cougarnet.uh.edu/pyuan2/Projects/LibFewShot/core/model/backbone/conv_four.py", line 69, in forward
    out1 = self.layer1(x)
  File "/home/cougarnet.uh.edu/pyuan2/anaconda3/envs/py37/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/cougarnet.uh.edu/pyuan2/anaconda3/envs/py37/lib/python3.7/site-packages/torch/nn/modules/container.py", line 117, in forward
    input = module(input)
  File "/home/cougarnet.uh.edu/pyuan2/anaconda3/envs/py37/lib/python3.7/site-packages/torch/nn/modules/module.py", line 727, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/cougarnet.uh.edu/pyuan2/Projects/LibFewShot/core/model/backbone/utils/maml_module.py", line 63, in forward
    if self.weight.fast is not None and self.bias.fast is not None:
AttributeError: 'Tensor' object has no attribute 'fast'

Thank you very much! The following documents are missing ： FileNotFoundError: [Errno 2] No such file or directory: './config/backbones/Conv64FLeakyReLU.yaml'

Traceback (most recent call last): File "E:/code/mildew/LibFewShot/run_trainer.py", line 31, in trainer = Trainer(config) TypeError: init() missing 1 required positional argument: 'config'

你好，我在Protonet网络中计算距离矩阵时乘以了0.01，发现训练准确率提升了，请问这个是正常的吗？还是说最终的准确率与距离的数量级也有关系。 "euclidean": lambda x, y: -0.01*torch.sum( torch.pow(x.unsqueeze(2) - y.unsqueeze(1), 2), dim=3,

I was trying to reproduce some results from the paper, and had no problem training CAN with a Conv64 backbone using the can.yaml file. However, when using Resnet12 as the backbone I had to lower episode_size to 2 (otherwise I'd get a CUDA out of memory error), and after that I couldn't figure out a working configuration to get the training started. Can someone help me out?

config file:

includes:
  - headers/data.yaml
  - headers/device.yaml
  - headers/misc.yaml
  - headers/model.yaml
  - headers/optimizer.yaml
  - classifiers/CAN.yaml
  - backbones/Conv64F.yaml

episode_size: 2

classifier:
  name: CAN
  kwargs:
    scale_cls: 7 # default
    num_class: 64
    nFeat: 64
    HW: 10

# backbone:
#   name: Conv64F
#   kwargs:
#     is_flatten: False
#     is_feature: False
#     leaky_relu: True
#     negative_slope: 0.2
#     last_pool: False

backbone:
  name: resnet12
  kwargs:
    maxpool_last2: False
    is_flatten: False

# need to modify: core/model/backbone/resnet_18.py:114 stride=1
# backbone:
#   name: resnet18
#   kwargs:
#     is_flatten: False
#     avg_pool: False

#backbone:
#  name: WRN
#  kwargs:
#    depth: 28
#    widen_factor: 10
#    avg_pool: False
#    is_flatten: False

log:

[11/18/21 12:49:40] INFO     {'data_root':                         trainer.py:50
                             '/storage/miniImageNet--ravi',
                             'image_size': 84, 'use_memory':
                             False, 'augment': True,
                             'augment_times': 1,
                             'augment_times_query': 1,
                             'device_ids': 0, 'n_gpu': 1, 'seed':
                             0, 'deterministic': True, 'log_name':
                             None, 'log_level': 'info',
                             'log_interval': 100,
                             'log_paramerter': False,
                             'result_root': './results',
                             'save_interval': 10, 'save_part':
                             ['emb_func'], 'tag': None, 'epoch':
                             50, 'test_epoch': 5, 'parallel_part':
                             ['emb_func'], 'pretrain_path': None,
                             'resume': False, 'way_num': 5,
                             'shot_num': 1, 'query_num': 15,
                             'test_way': 5, 'test_shot': 1,
                             'test_query': 15, 'episode_size': 2,
                             'train_episode': 100, 'test_episode':
                             100, 'batch_size': 128, 'optimizer':
                             {'name': 'Adam', 'kwargs': {'lr':
                             0.001}, 'other': {'emb_func':
                             0.001}}, 'lr_scheduler': {'name':
                             'StepLR', 'kwargs': {'gamma': 0.5,
                             'step_size': 10}}, 'includes':
                             ['headers/data.yaml',
                             'headers/device.yaml',
                             'headers/misc.yaml',
                             'headers/model.yaml',
                             'headers/optimizer.yaml',
                             'classifiers/CAN.yaml',
                             'backbones/Conv64F.yaml'],
                             'classifier': {'name': 'CAN',
                             'kwargs': {'scale_cls': 7,
                             'num_class': 64, 'nFeat': 64, 'HW':
                             10}}, 'backbone': {'name':
                             'resnet12', 'kwargs':
                             {'maxpool_last2': False,
                             'is_flatten': False}}, 'tb_scale':
                             1.0}
                    INFO     CAN(                                 trainer.py:355
                               (emb_func): ResNet(
                                 (layer1): Sequential(
                                   (0): BasicBlock(
                                     (conv1): Conv2d(3, 64,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn1): BatchNorm2d(64,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (relu):
                             LeakyReLU(negative_slope=0.1)
                                     (conv2): Conv2d(64, 64,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn2): BatchNorm2d(64,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (conv3): Conv2d(64, 64,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn3): BatchNorm2d(64,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (maxpool):
                             MaxPool2d(kernel_size=2, stride=2,
                             padding=0, dilation=1,
                             ceil_mode=False)
                                     (downsample): Sequential(
                                       (0): Conv2d(3, 64,
                             kernel_size=(1, 1), stride=(1, 1),
                             bias=False)
                                       (1): BatchNorm2d(64,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     )
                                     (DropBlock): DropBlock()
                                   )
                                 )
                                 (layer2): Sequential(
                                   (0): BasicBlock(
                                     (conv1): Conv2d(64, 160,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn1): BatchNorm2d(160,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (relu):
                             LeakyReLU(negative_slope=0.1)
                                     (conv2): Conv2d(160, 160,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn2): BatchNorm2d(160,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (conv3): Conv2d(160, 160,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn3): BatchNorm2d(160,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (maxpool):
                             MaxPool2d(kernel_size=2, stride=2,
                             padding=0, dilation=1,
                             ceil_mode=False)
                                     (downsample): Sequential(
                                       (0): Conv2d(64, 160,
                             kernel_size=(1, 1), stride=(1, 1),
                             bias=False)
                                       (1): BatchNorm2d(160,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     )
                                     (DropBlock): DropBlock()
                                   )
                                 )
                                 (layer3): Sequential(
                                   (0): BasicBlock(
                                     (conv1): Conv2d(160, 320,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn1): BatchNorm2d(320,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (relu):
                             LeakyReLU(negative_slope=0.1)
                                     (conv2): Conv2d(320, 320,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn2): BatchNorm2d(320,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (conv3): Conv2d(320, 320,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn3): BatchNorm2d(320,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (maxpool):
                             MaxPool2d(kernel_size=2, stride=2,
                             padding=0, dilation=1,
                             ceil_mode=False)
                                     (downsample): Sequential(
                                       (0): Conv2d(160, 320,
                             kernel_size=(1, 1), stride=(1, 1),
                             bias=False)
                                       (1): BatchNorm2d(320,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     )
                                     (DropBlock): DropBlock()
                                   )
                                 )
                                 (layer4): Sequential(
                                   (0): BasicBlock(
                                     (conv1): Conv2d(320, 640,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn1): BatchNorm2d(640,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (relu):
                             LeakyReLU(negative_slope=0.1)
                                     (conv2): Conv2d(640, 640,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn2): BatchNorm2d(640,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (conv3): Conv2d(640, 640,
                             kernel_size=(3, 3), stride=(1, 1),
                             padding=(1, 1), bias=False)
                                     (bn3): BatchNorm2d(640,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     (maxpool):
                             MaxPool2d(kernel_size=2, stride=2,
                             padding=0, dilation=1,
                             ceil_mode=False)
                                     (downsample): Sequential(
                                       (0): Conv2d(320, 640,
                             kernel_size=(1, 1), stride=(1, 1),
                             bias=False)
                                       (1): BatchNorm2d(640,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                     )
                                     (DropBlock): DropBlock()
                                   )
                                 )
                                 (avgpool):
                             AvgPool2d(kernel_size=5, stride=1,
                             padding=0)
                                 (dropout): Dropout(p=0.0,
                             inplace=False)
                               )
                               (cam_layer): CAMLayer(
                                 (cam): CAM(
                                   (conv1): ConvBlock(
                                     (conv): Conv2d(100, 10,
                             kernel_size=(1, 1), stride=(1, 1))
                                     (bn): BatchNorm2d(10,
                             eps=1e-05, momentum=0.1,
                             affine=True,
                             track_running_stats=True)
                                   )
                                   (conv2): Conv2d(10, 100,
                             kernel_size=(1, 1), stride=(1, 1))
                                 )
                                 (classifier): Conv2d(64, 64,
                             kernel_size=(1, 1), stride=(1, 1))
                               )
                               (loss_func): CrossEntropyLoss(
                                 (logsoftmax): LogSoftmax(dim=1)
                               )
                             )
                    INFO     Trainable params in the model:       trainer.py:357
                             12430610
[11/18/21 12:49:45] INFO     load 38400 image with 64 label.       dataset.py:94
                    INFO     load 9600 image with 16 label.        dataset.py:94
                    INFO     load 12000 image with 20 label.       dataset.py:94
                    INFO     Adam (                               trainer.py:447
                             Parameter Group 0
                                 amsgrad: False
                                 betas: (0.9, 0.999)
                                 eps: 1e-08
                                 initial_lr: 0.001
                                 lr: 0.001
                                 weight_decay: 0

                             Parameter Group 1
                                 amsgrad: False
                                 betas: (0.9, 0.999)
                                 eps: 1e-08
                                 initial_lr: 0.001
                                 lr: 0.001
                                 weight_decay: 0
                             )
                    INFO     ============ Train on the train set   trainer.py:67
                             ============
Traceback (most recent call last):
  File "/src/run_trainer.py", line 12, in <module>
    trainer.train_loop()
  File "/src/core/trainer.py", line 68, in train_loop
    train_acc = self._train(epoch_idx)
  File "/src/core/trainer.py", line 143, in _train
    output, acc, loss = self.model.set_forward_loss(batch)
  File "/src/core/model/metric/can.py", line 413, in set_forward_loss
    support_feat, query_feat, support_targets_one_hot, query_targets_one_hot
  File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/src/core/model/metric/can.py", line 214, in forward
    prototypes, query_feat
  File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/src/core/model/metric/can.py", line 149, in forward
    a1 = self.get_attention(a1)  # [1, 5, 75, 36]
  File "/src/core/model/metric/can.py", line 119, in get_attention
    a = F.relu(self.conv1(a))
  File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/src/core/model/metric/can.py", line 92, in forward
    return self.bn(self.conv(x))
  File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/module.py", line 1102, in _call_impl
    return forward_call(*input, **kwargs)
  File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 446, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/opt/conda/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 443, in _conv_forward
    self.padding, self.dilation, self.groups)
RuntimeError: Given groups=1, weight of size [10, 100, 1, 1], expected input[2, 289, 75, 5] to have 100 channels, but got 289 channels instead