{
  "id": 92177,
  "title": "keep going...",
  "url": "/competitions/human-protein-atlas-image-classification/discussion/92177",
  "author_name": "Wei Wang",
  "post_date": "2019-05-14T06:25:18.084000",
  "votes": 0,
  "comment_count": 0,
  "views": 0,
  "content": "<p>In this <a href=\"https://github.com/wangjiusi/Human-Protein-Atlas-Image-Classification\"><em>notebook</em></a>, I want to explore as much information as possible from raw data (without external data). And refer to <a href=\"https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb\"><em>https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb</em></a> and <a href=\"https://www.kaggle.com/hortonhearsafoo/fastai-v1-starter-pack-kernel-edition-lb-0-323\"><em>Fastai v1 starter pack (Kernel edition)</em></a>, I learned a lot from them and great thanks. Eventually, I get 0.47 (public score). By the way, no focal loss function!</p>",
  "messages": [
    {
      "id": 531022,
      "postDate": "2019-05-14T06:25:18.083Z",
      "content": "<p>In this <a href=\"https://github.com/wangjiusi/Human-Protein-Atlas-Image-Classification\"><em>notebook</em></a>, I want to explore as much information as possible from raw data (without external data). And refer to <a href=\"https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb\"><em>https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb</em></a> and <a href=\"https://www.kaggle.com/hortonhearsafoo/fastai-v1-starter-pack-kernel-edition-lb-0-323\"><em>Fastai v1 starter pack (Kernel edition)</em></a>, I learned a lot from them and great thanks. Eventually, I get 0.47 (public score). By the way, no focal loss function!</p>",
      "rawMarkdown": "In this [*notebook*](https://github.com/wangjiusi/Human-Protein-Atlas-Image-Classification), I want to explore as much information as possible from raw data (without external data). And refer to [*https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb*](https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb) and [*Fastai v1 starter pack (Kernel edition)*](https://www.kaggle.com/hortonhearsafoo/fastai-v1-starter-pack-kernel-edition-lb-0-323), I learned a lot from them and great thanks. Eventually, I get 0.47 (public score). By the way, no focal loss function!"
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "531022": "In this [*notebook*](https://github.com/wangjiusi/Human-Protein-Atlas-Image-Classification), I want to explore as much information as possible from raw data (without external data). And refer to [*https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb*](https://www.kaggle.com/iafoss/pretrained-resnet34-with-rgby-0-460-public-lb) and [*Fastai v1 starter pack (Kernel edition)*](https://www.kaggle.com/hortonhearsafoo/fastai-v1-starter-pack-kernel-edition-lb-0-323), I learned a lot from them and great thanks. Eventually, I get 0.47 (public score). By the way, no focal loss function!"
  }
}