-
# Background
This document serves as a plan for the final publication in my PhD. This work is a continuation of the novelty-based detection scheme proposed in the last two publication. Here we discus…
-
Hello, first of all, thank you for your great work.
I just inserted the self-attention and cross-attention module to stylegan2-ada, following behind the conv1 SynthesisLayer on the 4,8,16,32 resolut…
-
Hi, I'm implenting code in Google Colab. And I find a strange error.
```python
import torch
from musiclm_pytorch import MuLaN, AudioSpectrogramTransformer, TextTransformer
audio_transformer = …
-
Hi, thank you for sharing the code! It's really helpful!
I am trying to reproduce the performance on the VAST dataset, I am using python 3.6, Pytorch 1.10.1, cuda 11.3. I tried to use the same hype…
-
I have a question about codes for negative key =None:
If negative key is None, then Negative keys are implicitly off-diagonal positive keys.
# Cosine between all combinations
logits = que…
-
It seems that the contrastive loss have no gradient on the network. The "gen_img_emb" is generated by fixed CLIP while the " psed_emb" is pre-computed in dataloader:
```
with torch.no_grad():
…
-
### Describe the bug
when running the oobabooga fork of GPTQ-for-LLaMa, after about 28 replies a CUDA OOM exception is thrown.
```
torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to alloca…
-
Traceback (most recent call last):
File "/Users/introvertuoso/PycharmProjects/SpokenDigits/model_neural/contrastive_training.py", line 22, in
_, _ = contrastive_train(model, train_set, val_set,…
-
Hi,
I really liked your paper and I was trying to reproduce your results, so I highly appreciate that you shared the implementation. I have a question regarding the general proposed approach:
How ar…
-
I ran the program on the V100 using the script below, but it takes 42 minutes to run one epoch, I want to ask how long it took you to run the model?
```python main.py --batch_size 16 --plm bert --con…