-
### Reason/inspiration (optional)
Pages in docs/content/ai/concepts/neural-networks/terms including backpropagation.md, activation-function.md, binary-activation-function.md, sigmoid-activation-funct…
-
Primero agradecerte por el programa para iniciar con el aprendizaje de las redes neurales en Python. El error que obtengo es evidentemente es de principiante, he intentado solucionarlo y nada que pued…
-
Approximately 30% of all random neural network weight initializations cannot be trained with backpropagation (or at least the current implementation of backpropagation) when using nested neural networ…
-
It seems that the word embedding are kept static during training.
How to make the embedding changeable in backpropagation?
-
### Question
Hello, I am currently utilizing LLaVA1.5, which comprises both text-only and image-text instructions, for instruction fine-tuning within the ZERO-3 framework. However, I've encountered a…
-
您好,我在运行您的代码后,程序在迭代一轮之后,进入了死循环,如下图:
![image](https://github.com/user-attachments/assets/293c3c14-c669-48ae-9e42-5a78dd98b830)
🎈在进入iter5之后,通过如下代码就进入了死循环:
![image](https://github.com/user-attachme…
-
Thank you for your great work!
I ran your optimization code and found that the reason for the failure of optimization was that loss did not have backpropagation, but only needed just add a loss.back…
-
I use multiple A6000 cards for pretraining. The RAM of each card is 49140MiB.
I tried to pretrain LLaMA-7B with `bf16-mixed`,
```
batch_size = 60 # 125
micro_batch_size = 1 # 1 × 4 = 4 for each…
-
Dear @aadityacs,
It is nice to meet you here. I have been studying your paper on "TOuNN: Topology Optimization using Neural Networks" and I have a question regarding the update of network parameter…
-
Cool project! Where (and how) do you compute backpropagation in the multiresolution encoding?