Dipkumar Patel

Bounding box crop

8

## Description This PR adds feature requests of Issue #5428. `crop` supports cropping of ndarray on arbitrary axis given by the user. Image data is sliced/cropped based on coordinates given...

Filename : pricing/price-optimization-using-dqn-reinforcement-learning.ipynb at t=0, the Below function evaluates with p[0] and p[-1] as parameters which seems incorrect to me. because p[-1] in python corresponds to last element of the...

Experiment with lemmatization and/or stemming to improve search rankings

1

https://github.com/mwmbl/mwmbl/blob/02bcef640cfdc48f561e09f54e3ec2634b2e19fa/mwmbl/indexer/index.py#L31 Instead of indexing on lowercase tokens. We can use either lemmatization or stemming. (https://stackoverflow.com/questions/1787110/what-is-the-difference-between-lemmatization-vs-stemming) Spacy lemmatizer: https://spacy.io/api/lemmatizer Pros: Will have reduced vocabulary ({play, playing, played} => {play}) Will improve...

I am trying to reproduce result in PyTorch but loss already start from 10e-2. ``` class ModMSELoss(torch.nn.Module): def __init__(self,shape_r_gt,shape_c_gt): super(ModMSELoss, self).__init__() self.shape_r_gt = shape_r_gt self.shape_c_gt = shape_c_gt def forward(self, output...

feat: added k, v cache for inference speed up

4

Hi, @jaymody, Awesome blog post. I was interested in learning kvcache during inference and searched for it but existing articles on kvcache don't focus on the implementation part of it....

(enhancement) consider to integrate steganography

2

steganography will provide hidding data and and also some good amount of space utilization.