Abhishree Thittenamane

Upgrade to PTL 2.1

45

# What does this PR do ? Add a one line overview of what this PR aims to accomplish. **Collection**: [Note which collection this PR will affect] # Changelog -...

Handle float limit_val_batches

10

# What does this PR do ? 1) When a float `limit_val_batches` is passed by the user, the PR ensures that the `limit_val_batches` is cast to an equivalent int value...

# What does this PR do ? Add a one line overview of what this PR aims to accomplish. **Collection**: [Note which collection this PR will affect] # Changelog -...

Add nemo tests to ecosystem

1

## What does this PR do? Fixes # (add a link to the created issue in your repository if any) OR link your project for clarity. Before submitting - [...

Add support for distributed optimizer

5

Enable no_copy argument in c10d files for distributed optimizers

Implement TensorBase.div_

1

### 🚀 Feature Implement [TensorBase.div_](https://pytorch.org/docs/stable/generated/torch.Tensor.div_.html) ### Motivation NeMo's [MegatronNeVaModel](https://github.com/NVIDIA/NeMo/blob/main/examples/multimodal/multimodal_llm/neva/neva_pretrain.py) cc @tfogal

Implement TensorBase.is_complex

1

### 🚀 Feature Implement [`torch.is_complex`](https://pytorch.org/docs/stable/generated/torch.is_complex.html) ### Motivation NeMo's [MegatronNeVaModel](https://github.com/NVIDIA/NeMo/blob/main/examples/multimodal/multimodal_llm/neva/neva_pretrain.py) cc @apaz-cli @tfogal

Implement torch.device

2

### 🚀 Feature Implement [`torch.device`](https://pytorch.org/docs/stable/tensor_attributes.html#torch.device:~:text=float_tensor%20*%3D%20complex_float_tensor-,torch.device,-CLASS) ### Motivation NeMo [Vision Transformer](https://github.com/NVIDIA/NeMo/blob/main/examples/vision/vision_transformer/megatron_vit_classification_pretrain.py) cc @tfogal

Implement _VariableFunctionsClass.randint of torch

5

## 🚀 Feature Implement [`_VariableFunctionsClass.randint`](https://pytorch.org/docs/stable/generated/torch.randint.html) ### Motivation NeMo [Stable Diffusion](https://github.com/NVIDIA/NeMo/blob/main/examples/multimodal/text_to_image/stable_diffusion/sd_train.py) cc @tfogal

Implement _set_grad_enabled of torch._C

2

### 🚀 Feature Implement _set_grad_enabled of torch._C ### Motivation NeMo [Vision Transformer](https://github.com/NVIDIA/NeMo/blob/main/examples/vision/vision_transformer/megatron_vit_classification_pretrain.py) cc @tfogal