-
### Version Checks (indicate both or one)
- [ ] I have confirmed this bug exists on the lastest [release](https://github.com/pypsa/pypsa/releases) of PyPSA.
- [ ] I have confirmed this bug exists on…
-
### System Info
```
(zt) root@autodl-container-7071118252-7032359d:~/test/PiPPy/examples/llama# transformers-cli env
Copy-and-paste the text below in your GitHub issue and FILL OUT the two last p…
-
Traceback (most recent call last):
File "train.py", line 135, in
test_abs(args, device_id, cp, step)
File "E:\project\PreSumm\src\train_abstractive.py", line 215, in test_abs
model = …
-
**Is your feature request related to a problem? Please describe.**
Would be great to be able to load a LoRA to a model compiled with `torch.compile`
**Describe the solution you'd like.**
Do `load…
-
### Describe your use-case.
Flux has layers named single_transformer_blocks.* and transformer_blocks.*.
If I want to train only the **transformer_blocks.*** layers but exclude **single_transformer…
-
### 🚀 The feature, motivation and pitch
I am trying to extract hidden states from the final layer of llama3-8b (i.e., the final batch_size, seq_length, n_emb vector _before_ computing the logits). Wo…
-
### Describe the workflow you want to enable
I want to be able to use multiple estimators in one pipeline. E.g.
```python
from sklearn.pipeline import Pipeline
from sklearn.linear_model impor…
-
When I load the existing pretrained model, the following error was reported: "RuntimeError: Error(s) in loading state_dict for FairModel4CIKM:
Missing key(s) in state_dict: "i_embeddings.weight", "p…
-
**Describe the bug**
There is a mismatch between the `train.yaml` configuration file and the loaded model weights (`final.pt`) when using the Wenet pretrained model `wenetspeech_u2pp_conformer_exp`. …
-
Here is what I am getting (see below)
FP8 slower than FP16
for FP16, multiples of 16 make things slower than multiple of 8
Am I missing something ?
Batch_size_multiple 16 // Seqlen multi…