Q1: How do I clone repo in my drive?
Mount the drive using:
from google.colab import drive
drive.mount('/content/drive/')
Clone the repo in your drive
%cd /content/drive/<name of your drive>
!git clone <github repo url>
Q2: How do I use Colab?
Resource:
Q3: I am trying to run the codes on Colab but got AttributeError: module 'PIL.Image' has no attribute 'register_decoder'.
Execute !pip install Pillow==4.1.1 or !pip install --no-cache-dir -I pillow. Restart your notebook.
Q4: In Colab do I need to write the lines related to cuda?
Yes.
Q5: How can I get the cat-dog dataset? Run the following commands:
!wget "https://s3.amazonaws.com/content.udacity-data.com/nd089/Cat_Dog_data.zip" -P "pytorch_challenge/transfer_learning"
!unzip -qq -o "pytorch_challenge/transfer_learning/Cat_Dog_data.zip" -d "pytorch_challenge/transfer_learning"`
!ls "/content/pytorch_challenge/transfer_learning/Cat_Dog_data"
data_dir = "/content/pytorch_challenge/transfer_learning/Cat_Dog_data"
Q6: How can I save bandwidth at the expense of drive Storage when training my model?
From @ecdrid
With your filename as xyz.pth, after some training:
# This only needs to be done once in a notebook.
!pip install -U -q PyDrive
from pydrive.auth import GoogleAuth
from pydrive.drive import GoogleDrive
from google.colab import auth
from oauth2client.client import GoogleCredentials
# Which file to send?
file_name = "xyz.pth" #make sure you always change this..
from googleapiclient.http import MediaFileUpload
from googleapiclient.discovery import build
auth.authenticate_user()
drive_service = build('drive', 'v3')
def save_file_to_drive(name, path):
file_metadata = {'name': name, 'mimeType': 'application/octet-stream'}
media = MediaFileUpload(path, mimetype='application/octet-stream', resumable=True)
created = drive_service.files().create(body=file_metadata, media_body=media, fields='id').execute()
return created
save_file_to_drive(file_name, file_name)
Now you can just connect your drive and start training further if you wish.