-
Do you have plans to integrate weights and biases?
-
Hi,
I ran a simple simulation to assess IVON's performance in recovering confidence intervals with appropriate coverage. Its a simple linear regression problem, which you will find below.
```
…
-
### Describe the bug
When using wandb client or the API to download artifact from wandb i get the following error:
wandb artifact get sagemaker-demo/cifar10_dataset
wandb: Downloading dataset artifa…
-
We have ops `prepare_conv_weights_for_ttnn` and `prepare_conv_bias_for_ttnn` that will pre-format convolution weights and biases for conv2d (without pushing them to device). In the previous implementa…
-
Users can already use weights and biases in their Flyte workflows. For example, we have a tutorial that uses [W&B for metrics tracking](https://docs.flyte.org/projects/cookbook/en/latest/auto/case_stu…
-
My organization only allows pytroch models to go into production. I wrote the following code to convert my trained h2o model into pytorch but it errors out.
```
import torch
import torch.nn as nn…
-
When performing convolution calculations, errors are not supported if the weights and biases are discontinuous.
![image](https://github.com/user-attachments/assets/47078eef-bc88-4da3-a283-bebb9f04953…
LRJKD updated
1 month ago
-
I downgraded to Keras 2 hoping to use lwtnn after reading this issue: https://github.com/lwtnn/lwtnn/issues/175
Unfortunately it still doesn't work. The issue is that lwtnn expects a specific struc…
-
Currently there are quite some logics to decide how a weight tensor should be put on device: [prepare_conv_weights_biases_and_move_to_device](https://github.com/tenstorrent/tt-metal/blob/14a6f6f086e88…
-
**Describe the bug**
When I run the example from examples/python/awq-quantized-model.md, but switching out phi-3 for llama-3.2-3b, I get an error message stating that `AttributeError: 'NoneType' objec…