Add image encoder base #61
Conversation
EdenWuyifan
force-pushed
the
image_encoder
branch
3 times, most recently
from
e61ac96
to
dc57a9d
Compare
|
|
||
| import numpy as np | ||
| import pandas as pd | ||
| from skimage.color import gray2rgb, rgb2gray, rgba2rgb |
There was a problem hiding this comment.
The skimage lib should be loaded as an optional dependency, otherwise it will raise an error because it's not installed in the default version of alpha-automl
alpha_automl/resource/Untitled.ipynb
Outdated
There was a problem hiding this comment.
This file shouldn't be in the resource folder
There was a problem hiding this comment.
This file is not necessary
There was a problem hiding this comment.
Remove the d3mIndex column. Avoid to use the name learningData.csv. Use something more descriptive, e.g. digits.csv, etc. It's not necessary to have the folder tables. Change the name image_dataset_2 to digits
There was a problem hiding this comment.
Add only one example for image classification. Add the link to download the dataset in the jupyter notebook. This link should contain the trian_data.csv and test_data.csv files (use the same names that we have in the other example to be consistent). Avoid to split the data in the jupyter notebook. The less code we put in our example, the better.
EdenWuyifan
force-pushed
the
image_encoder
branch
2 times, most recently
from
3a929ba
to
8dbd6b4
Compare
…CLIP; update for semi supervised tasks; add selfie notebook; add source for dataset; add testcases
EdenWuyifan
force-pushed
the
image_encoder
branch
from
8dbd6b4
to
ea19fa1
Compare
|
I made some minor changes. Feel free to merge it. |
|
I fixed some bugs and new it is ready to merge |
No description provided.