Yash Akhauri

Check out the link here: https://leonardoaraujosantos.gitbooks.io/artificial-inteligence/content/making_faster.html Now, for better understanding: open im2col.py and add the code below at the bottom. Then run python im2col.py and observe the output: ``` init...

Alternatively, just change this function to: ``` def im2col_batch(input, kernel_size, stride, padding): if input.dim() == 3: return _im2col(input, kernel_size, stride, padding) elif input.dim() == 4: shape = (input.size(0),) + im2col_shape(input.size()[1:],...

I do not mind implementing ResNet for this if there is interest. However, I will not be able to train it. I think this task mostly just involves changing the...

@XinDongol could you kindly share the python models directory code for the ResNet-xNOR with me? Thanks!

Hello. I have been implementing this in tensorflow with reference to [this](https://github.com/jonrei/tf-AdaIN) . I am getting results with *VERY* low contrast, as in the pictures are extremely dull. Implementing the...

Hello, Has there been a resolution / more discussion on this? I think the simple fix is to do this [here](https://github.com/FasterDecoding/SnapKV/blob/main/snapkv/monkeypatch/llama_hijack_4_37.py#L90C13-L90C40): `key_states, value_states = past_key_value.update(key_states_compress, value_states_compress, self.layer_idx, cache_kwargs) ` Thanks!

Still experiencing this issue on A6000 with 0.7.3 when requesting `logprobs` with `echo=True` Tried with/without prefix-caching and chunked-prefill.