-
Will you provide requirements.txt?
-
**Describe the bug**
When try to use rknn toolkit to build the rknn model from pytorch efficientNet model :
from rknn.api import RKNN
trace_model = torch.jit.trace(model,input)
trace_model.save(…
-
Pollutes `r(max)` in SEResNets, and `r(min)` isn't sufficiently informative
Meant to open this as feature request
-
Thanks for your error report and we appreciate it a lot.
**Checklist**
1. I have searched related issues but cannot get the expected help.
2. I have read the [FAQ documentation](https://mmengin…
-
## 🐛 Bug
The current 0.6.0 is not able to load a fine tuned model file from local path
This is the link to the model file: https://syncortex-my.sharepoint.com/:u:/p/satish_jasthi/ET3rE5L3KZRDkC…
-
-
## Issue description
I am Randomly getting this runtime error in mid-epochs.
## Code example
```
RuntimeError: tensorflow/compiler/xla/xla_client/xrt_computation_client.cc:421 : Check failed:…
-
### 🐛 Describe the bug
```python
>>> mytensor = torch.rand(40,32,256,256)
>>> torch.save(mytensor[0], "tmp.pt")
>>> newtensor = torch.load("tmp.pt")
>>> sys.getsizeof(mytensor.storage())
335…
-
### 🐛 Describe the bug
I attempted to run the BERT example on two GPUs in a single node using following command:
`torchrun --nproc_per_node 1 --master_addr localhost --master_port 29500 train.py`
…
-
**Describe the issue**:
When I don't write the code in the green box, it can't execute normally, and the webui shows that it takes 6 seconds to execute the successed, which is not right. But after I …