-
Hi all -
Hoping that there is someone out there who has figured out a solution to my problem.
I am trying to train a mean-teacher (https://github.com/CuriousAI/mean-teacher) network that uses bo…
-
I am using a customed dataset where the data is loaded from disk in `__init__` function of dataset. But I found that the data will be loaded n times if I use n gpus (which also means the `num_processe…
-
新装的环境,python 3.10
magic-pdf 0.6.1
detectron2 0.6
paddleocr 2.7.3
paddlepaddle 2.6.1
运行
magic-pdf pdf-command --pdf "test_pdf_files_…
-
Initialize and pre-train a smaller model
Please try initializing a smaller model
-
I am trying to train an object detection model using custom data without segmentation. However, I encountered the following error during the training process:
```python
Traceback (most recent call…
-
### Bug description
Running with the LightningCLI, MLflow logger, and `MLFLOW_TRACKING_URI` environment variable set causes an assertion failure with logging. I think using a remote tracking server…
-
Hi Nils, quick question, what is the best way to train with MNRL? For one query can I have a list of positive passages? Or should only have one passage? And can a query have many positive passages?
…
-
DeepChem has some limitations at present when featurizing large datasets. In particular, here are a couple issues that crop out:
- There's no out-of-the-box way to featurize a large collection of o…
-
### 🐛 Describe the bug
I am using a customized HeteroGraph Dataset(in PyG), and would like to mix this dataset with another different dataset(e.g. image).
To do that, I want to load data with re…
-
When i train with your config,i encounter a problem,it is :
Traceback (most recent call last):
File "src/tools/train.py", line 108, in
main()
File "src/tools/train.py", line 86, in main
…