Omri Kaduri
Omri Kaduri
In `split_label.py` file, you have a broken import Try change this: ```python from utils.prepare.utils import orderConvex, shrink_poly ``` to ```python from utils import orderConvex, shrink_poly ```
Hi, Your example of using the U-Net uses 'same' convolutions. How should I prepare my data for 'valid' convolutions? I use the [ISBI dataset](http://brainiac2.mit.edu/isbi_challenge/), which input size (and label size)...
Link to page(s) with issue: https://planning.wiki/guide/whatis/aip Quote of current error: "Therefore we need an appraoch which is methodical" Proposed correction: Therefore we need an approach which is methodical
Hi @karpathy, thanks for that great repo! Maybe it would be better to note in your code that while you're training by [minimizing the CE loss](https://github.com/karpathy/makemore/blob/f61811b994280cb12ddae15ef5800baa2e3a1ca4/makemore.py#L392), Bengio actually **maximized** the...
Hey, I've wanted to use your classifier for a multiclass problem. The easiest way I've thought about is using sklearn `OneVsRestClassifier` and refactoring your `Classifier` to be an sklearn estimator...
Pixtral-12B tokenizer error - special_token_policy=IGNORE does not ignore special tokens in decoding
### Python -VV ```shell Decoding `tokens` that contain special tokens ([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,...