yang-zhang

yang-zhang

tar -xf archive.tar -C /target/directory

wget url -P /path/to/folder

gunzip -c file.gz > /THERE/file

unzip source.zip -d /path/to/folder

yang-zhang

def sample_dct(dct, n=3):
    idx = np.random.choice(len(dct), n)
    return dict(np.array(list(dct.items()))[idx])

yang-zhang

• ModuleList: https://pytorch.org/docs/stable/nn.html#torch.nn.ParameterList

  class MyModule(nn.Module):
  def __init__(self):
      super(MyModule, self).__init__()
      self.linears = nn.ModuleList([nn.Linear(10, 10) for i in range(10)])
  
  def forward(self, x):
      # ModuleList can act as an iterable, or be indexed using ints
  

for i, l in enumerate(self.linears):

yang-zhang

• paper doll
  • has dataset
  • KNN
• Creating Capsule Wardrobes from Fashion Images
• Learning Fashion Compatibility with Bidirectional LSTMs
• Parsing Clothing in Fashion Photographs
  • we provide a large novel dataset and tools
• https://github.com/ihciah/deep-fashion-retrieval
• https://github.com/wenxinxu/deep-shopping
• https://github.com/leekyungmoon/WatchOUT

yang-zhang

extending

https://pytorch.org/docs/master/notes/extending.html

pytorch_summary

https://github.com/sksq96/pytorch-summary/blob/master/torchsummary/torchsummary.py

PyTorch Tutorial for Deep Learning Researchers

• https://github.com/yunjey/pytorch-tutorial/tree/master/tutorials
• 6k stars

pytorch/examples

yang-zhang

while else # nobreak

yang-zhang

Notes on fast.ai machine learning course by Jeremy Horward

• video: https://youtu.be/CzdWqFTmn0Y
• github: https://github.com/fastai/fastai/tree/master/courses/ml1

Lesson 1

• pd.read_csv(low_memory=False, parse_dates) https://youtu.be/CzdWqFTmn0Y?t=1717
• display_all function for df: https://youtu.be/CzdWqFTmn0Y?t=1929
• log error cares about ratio rather than value: https://youtu.be/CzdWqFTmn0Y?t=2107

yang-zhang

dict_ = {'001.jpg': 'dog', '002.jpg': 'cat', }
df = pd.DataFrame.from_dict(dict_, orient='index').reset_index()
df.columns = ['image_name', 'tags']

yang-zhang

pickle.dump(something, open('something.p', 'wb'))
something = pickle.load(open('something.p', 'rb'))