-
I downloaded sa_000000.tar from SA-1b to try to train masa. When I followed the tutorial to complete the dataset format conversion and train it, I got this error:
loading annotations into memory...
…
-
In my understanding, in pretrain code, it broadcasts the data from tp rank 0 to the rest tp rank gpus.
However, if i activate the option `train_valid_test_datasets_provider.is_distributed = True` wh…
-
- Add more elaborate description:
**gcube** is an R package that provides a simulation framework for biodiversity data cubes. This can start from simulating multiple species distributed in a landsc…
-
### Feature request
It would be useful to be able to straightforwardly repeat iterable datasets indefinitely, to provide complete control over starting and ending of iteration to the user.
An It…
-
I have a device containing 4 Nvidia L40 GPUs. I am trying to use the full_finetune_distributed llama3_1/8B_full recipe. My configuration for dataset in the config file is given below:
dataset:
_c…
-
**Describe the bug**
I encountered the error "OverflowError: int too big to convert" when trying to run `ilab model train` on my local system.
**To Reproduce**
Steps to reproduce the behavior:
1…
-
**Describe the bug**
The warning
```
2024-10-11 00:04:31,529 - distributed.worker.memory - WARNING - Unmanaged memory use is high. This may indicate a memory leak or the memory may not be released t…
-
### Describe the bug
The sharding of IterableDatasets with respect to distributed and dataloader worker processes appears problematic with significant performance traps and inconsistencies wrt to d…
-
# Description
When you load a dataset from HF with remote code, the load_dataset function prompts the user for permission to run remote code. This prompt only happens the first time the user downlo…
-
Hi, thank you for the great work!
The current implementation of the `PackedDataset` class only supports in-memory map-style datasets. When working with large datasets, the in-memory limitation can…