-
Hi everyone,
I quantized yolov3-tiny model with float16 and run the model in onnxruntime
```
ort_session = ort.SessionOptions()
ort_session.graph_optimization_level = ort.GraphOptimizationLevel.…
-
Hi,
while training the modeling, I always get this error (each time with a different pdb file).
As far as I understand, the TMP directory is created and the original pdb
file is moved there, but …
-
Test code:
```python
import torch
from functorch.compile import aot_function, make_boxed_func, min_cut_rematerialization_partition
def fn(x, y, z, a, b, c):
return (a * c) * x + (b) * y + …
-
Hello,
I am using g2o for 3D SLAM. In my application, the location of some of the landmarks (3D points) in the environment are known to me, for example I know the Euclidean distance between some of…
-
## Issue Description
In the course to find rooms for performance optimization, we are trying to go over the expensive calls in the consensus phases to improve performance.
This particular one …
-
Nearly all of the runtime (> 90%) now gets spent in ranking move candidates. This is because we currently rank move candidates based on the BIC score we get after performing the move and optimizing th…
-
I had a conversation with a person on Slack who tested out a the behavior of an LN connection to NGS with a user having `--deny-pubsub=">"` and enabling trace logs to discover the `[LS+] ...` lines wh…
bruth updated
7 months ago
-
**Is your feature request related to a problem? Please describe.**
I see quite some GPT-2 implementations, but I am missing a GPT-Neo/GPT-J implementation. This is a variant of GPT-2 which is quite s…
-
```
necktwi@CheapFellow:~/workspace/llm.c$ make train_gpt2cu USE_CUDNN=1 CUDNN_FRONTEND_PATH="/home/necktwi/workspace/cudnn-frontend/include"
necktwi@CheapFellow:~/workspace/llm.c$ ./train_gpt2cu …
-
### Describe the documentation issue
When I used Onnxruntime tvm for inference based on backend GPU and ROCM using a precompiled model, the following error occurred
Check failed: (itr != phy…