The only other requirements for this post is the basic understanding of the Python programming language, some experience working with Jupyter notebooks and having a Kaggle account which you can sign-up for here. The reason I am quite comfortable in writing this post is because slowly but surely, deep learning and more specifically, the tools for learning deep learning are becoming more and more accessible to the masses. I am a firm believer in not re-inventing the wheel when it is not necessary.
Therefore, the next step would be to follow along to this fantastic setup tutorial by Jeremy Howard who has been a guiding light for me in the realm of Deep Learning. Although the layout of the website has changed, the process is still the same. If you followed along, and got the same results as Jeremy, guess what?
You are now well and truly on the path to Deep Learning glory. Next, go to the terminal where you hopefully have your instance running. You may then SSH into your instance as Jeremy mentions in his video, using this command :. You should be able to see 4 directories — anaconda2, downloads, git and nbs. Our next step is to create a directory that stores our image data. The first order of business is to install the command line tool for Kaggle and that can be done using the command :.
What now? Yep, you guessed it! We download the competition data from Kaggle using the command :. Once the data is downloaded, you can see all the files that are downloaded. The two main files of importance are test.
Sign up or log in Sign up using Google. Sign up using Facebook. Sign up using Email and Password. Post as a guest Name. Email Required, but never shown.
The Overflow Blog. Podcast Who is building clouds for the independent developer? Exploding turkeys and how not to thaw your frozen bird: Top turkey questions Featured on Meta. Now live: A fully responsive profile. Reducing the weight of our footer. Jose Cherian Jose Cherian 5, 3 3 gold badges 32 32 silver badges 36 36 bronze badges. Awsome Thanks Jose Cherian — user I faced an issue — user You can check the implementation of the Kaggle API But if you are lazy you can just install kaggle on your server pip install kaggle.
And to download a whole competition you may call this from python. Thanks, I am aware of all that. Right now I am using the "lazy" solution of invoking os. Your comment does not tell anything that I didn't know or hadn't already tried, so no vote this time, I guess. So indeed, the best resource I have been able to find so far is the actual source for the api: github.
Sign up or log in Sign up using Google. Sign up using Facebook. Sign up using Email and Password. Post as a guest Name. Email Required, but never shown. Classification with Binary Weight Network.
Neural Network Compression. Image Recognition. Classification Consistency. Show all 79 benchmarks. Collapse benchmarks. Dataset Loaders Edit Add Remove. Tasks Edit. Similar Datasets. License Edit. Modalities Edit. Languages Edit. Contact us on: hello paperswithcode. Terms Data policy Cookies policy from. ImageNet ReaL. ImageNet 64x Local search. ImageNet 32x ImageNet V2. ImageNet x ImageNet 5 tasks. ImageNet Fine-grained 6 Tasks.
ImageNet ResNet - 90 Epochs. ImageNet 50 samples per class. ImageNet ResNet - 60 Epochs.
0コメント