Hello! I set up Salamander and was able to access the notebooks. I had no issues the first few times through, but I randomly started receiving the below error when trying to create a CNN using resnet34. Interestingly enough, I have no issues at all when I use resnet50. I’ve searched everywhere and can’t seem to find what may be causing this or how to fix it, and am a bit of a beginner so debugging the unpickling process is a bit tough at this point. Can anybody point me in the right direction?
Hello All,
This is my first attempt at any kind of Machine Learning. I wanted to discuss next steps after going through lesson1, apologies if a similar question has been asked before, feel free to point me to the answer if there exits one.
From what I understood the homework/ assignment for the lesson 1 is to try and build a image classification model on your own using your dataset.
Ques 1 : Is the correct ?
If i have a image classification model idea in mind lets say : Classify different modes of transport [bus, train, ship, airplanes] etc, my first step should be to create a dataset using one of the approaches mentioned in an answer(create dataset…). To create this, i should have some samples of each class and place them under train and valid data folders(if using that format). Once this is done, i could train a model and see the results.
I’m having a bit of trouble running the notebook. When I try to run
path = untar_data(URLs.PETS); path
I get an error spew in which the fundamental problem seems to be a failure to connect to the URL:
gaierror: [Errno -3] Temporary failure in name resolution
NewConnectionError: <urllib3.connection.VerifiedHTTPSConnection object at 0x7f8ec4ed7cf8>: Failed to establish a new connection: [Errno -3] Temporary failure in name resolution
I finished Lesson 1 and absolutely loved it.
I wanted to implement these techniques and functions elsewhere, and I tried to apply it to the MNIST Dataset that is available on Kaggle (not the one available in the fast.ai library).
I am stuck on the ImageDataBunch() function, and I am not sure what the “Path” argument should contain. I know how to import data from my system, and convert into a Data Frame or CSV and use the appropriate functions, but I am still not sure what my “Path” argument should contain.
Yep, went through the documentation. It seems to work only after all necessary arguments are added, such as valid_pct, for the from_folder type. Thanks for the help!
There is an easier way to download google images using the google-images-download python module in your jupyter notebook.
Here are the steps to download the different bears images
First install the python module !pip install google_images_download
Then in the subsequent cells, start downloadig the grizzly bears (-k grizzly argument below) !googleimagesdownload -k grizzly -l 100
A new folder called downloads/grizzly will be created with 100 images in it (that’s the -100 argument n the command above)
Do the same thing for the other categories !googleimagesdownload -k ‘black bear’ -l 100 !googleimagesdownload -k ‘teddy bear’ -l 100
Run the ls command !ls downloads/
and you should have the 3 following folders
‘black bear’ grizzly ‘teddy bear’
Hi earthyearth hope your are well!
From lesson 1 don’t think I heard a precise definition of a model.
However the post below gives one definition and is an easy read, combine it with lesson 1 info and it should help your understanding.
