Got it, let's tackle this—uploading an entire local dataset to Colab without manually picking each file is totally doable. Here are a few reliable methods I've used myself for different scenarios:

方法1：上传压缩包然后解压（适合中小数据集） #

This is the quickest fix if your dataset isn't massive. Instead of uploading dozens/hundreds of files, bundle them into one compressed file first:

• On your local machine, zip or tar.gz your entire dataset folder (right-click > "Compress" on most systems).
• In Colab, click the 📁 icon in the left sidebar, then hit the "Upload" button and select your compressed file.
• Once uploaded, run the corresponding command to extract it:
  • For zip files:

    !unzip your_dataset.zip
    

  • For tar.gz files:

    !tar -xzf your_dataset.tar.gz
    

• After running the command, you'll see your full dataset folder in the file browser—ready to use.

方法2：挂载Google Drive（适合大数据集/长期使用） #

If your dataset is large (10GB+) or you need to reuse it across multiple Colab sessions, this is the way to go. It avoids repeated uploads and is more stable:

• First, sync your local dataset folder to your Google Drive: Use the Google Drive desktop app (for seamless background sync) or drag the folder directly into your Drive via the web interface.
• In Colab, run this code to mount your Drive:

  from google.colab import drive
  drive.mount('/content/drive')
  

• Follow the prompt: Click the generated link, log into your Google account, copy the authorization code, paste it back into Colab's input box, and press Enter.
• Once mounted, your dataset will be accessible at /content/drive/MyDrive/[your_dataset_folder_path]—you can reference this path directly in your code.

方法3：直接上传整个文件夹（适合中小型 folders） #

Colab has a built-in method to upload entire folders without compression, though it's best for smaller folders (since upload speed depends on your network):

• Run this code in a Colab cell:

  from google.colab import files
  uploaded = files.upload_directory()
  

• A file picker will pop up—select your local dataset folder, and Colab will upload all its contents directly to the /content directory.

Quick Tips #

• Always verify the upload worked with !ls (lists files in the current directory) or !ls /content/drive/MyDrive (for Drive-mounted files).
• For extra large datasets, skip browser-based uploads entirely—use the Drive desktop app to sync locally first, then mount in Colab.

内容的提问来源于stack exchange，提问作者mohamed_abdullah