• Search by Module
• Search by Words
• Search Projects

• Java
• Python
• JavaScript
• TypeScript
• C++
• Scala
• Blog

• prediction-flow-master
  • .github
    • ISSUE_TEMPLATE
      • feature_request.md
      • bug_report.md
  • examples
    • movielens
      • ml-1m
        • README
        • preprocess.ipynb
      • movielens-1m.ipynb
    • amazon
      • prepare_neg.ipynb
      • amazon.ipynb
      • amazon-lightning.ipynb
  • LICENSE
  • CONTRIBUTING.md
  • setup.py
  • setup.cfg
  • .travis.yml
  • README.md
  • requirements.txt
  • .gitignore
  • MANIFEST.in
  • prediction_flow
    • transformers
      • __init__.py
      • column
        • category_encoder.py
        • log_transformer.py
        • sequence_encoder.py
        • standard_scaler.py
        • __init__.py
        • tests
          • test_standard_scaler.py
          • test_sequence_encoder.py
          • test_category_encoder.py
          • test_column_flow.py
          • __init__.py
          • test_log_transformer.py
        • column_flow.py
        • base.py
    • utils
      • __init__.py
    • pytorch
      • din.py
      • interest_net.py
      • wide_deep.py
      • deepfm.py
      • nn
        • interest.py
        • attention.py
        • pooling.py
        • fm.py
        • __init__.py
        • tests
          • test_fm.py
          • test_pooling.py
          • test_interest.py
          • test_rnn.py
          • __init__.py
          • test_mlp.py
          • test_attention.py
        • rnn.py
        • mlp.py
      • functions.py
      • dnn.py
      • data
        • __init__.py
        • tests
          • test_dataset.py
          • __init__.py
        • dataset.py
      • __init__.py
      • utils.py
      • tests
        • test_deepfm.py
        • test_dnn.py
        • test_din.py
        • __init__.py
        • utils.py
        • test_dien.py
        • test_wide_deep.py
      • dien.py
      • base.py
    • __init__.py
    • features
      • features.py
      • sequence_feature.py
      • number_feature.py
      • category_feature.py
      • __init__.py
      • tests
        • test_features.py
        • __init__.py
      • base.py
    • metrics
      • __init__.py

import torch.nn as nn
import torch.nn.init as init


def init_weights(model):
    if isinstance(model, nn.Linear):
        if model.weight is not None:
            init.kaiming_uniform_(model.weight.data)
        if model.bias is not None:
            init.normal_(model.bias.data)
    elif isinstance(model, nn.BatchNorm1d):
        if model.weight is not None:
            init.normal_(model.weight.data, mean=1, std=0.02)
        if model.bias is not None:
            init.constant_(model.bias.data, 0)
    elif isinstance(model, nn.BatchNorm2d):
        if model.weight is not None:
            init.normal_(model.weight.data, mean=1, std=0.02)
        if model.bias is not None:
            init.constant_(model.bias.data, 0)
    elif isinstance(model, nn.BatchNorm3d):
        if model.weight is not None:
            init.normal_(model.weight.data, mean=1, std=0.02)
        if model.bias is not None:
            init.constant_(model.bias.data, 0)
    else:
        pass