Awesome
Determined As-is Examples
This repository contains a variety of Determined examples that are not actively maintained by the Determined team.
Tutorials
| Example | Dataset | Framework |
|---|---|---|
| fashion_mnist_tf_keras | Fashion MNIST | TensorFlow (tf.keras) |
Blog posts
| Example | Description |
|---|---|
| Activation Memory: Part 2 | Measuring activation memory in PyTorch. |
| LLM Finetuning | Finetuning TinyLlama-1.1B on Text-to-SQL. |
| LLM Finetuning 2 | Finetuning Mistral-7B on Text-to-SQL using LoRA and DeepSpeed. |
| LLM Finetuning 3 | Finetuning Gemma-2B using DPO. |
| LoRA Parameters | Finding the best LoRA parameters. |
| Python SDK demo | Example usage of the Determined Python SDK to run and administer experiments. |
| Tensor Parallelism | Profiling tensor parallelism in PyTorch. |
Computer Vision
| Example | Dataset | Framework |
|---|---|---|
| cifar10_pytorch | CIFAR-10 | PyTorch |
| cifar10_pytorch_inference | CIFAR-10 | PyTorch |
| cifar10_tf_keras | CIFAR-10 | TensorFlow (tf.keras) |
| fasterrcnn_coco_pytorch | Penn-Fudan Dataset | PyTorch |
| mmdetection | COCO | PyTorch |
| detr_coco_pytorch | COCO | PyTorch |
| deformabledetr_coco_pytorch | COCO | PyTorch |
| iris_tf_keras | Iris Dataset | TensorFlow (tf.keras) |
| unets_tf_keras | Oxford-IIIT Pet Dataset | TensorFlow (tf.keras) |
| efficientdet_pytorch | COCO | PyTorch |
| byol_pytorch | CIFAR-10 / STL-10 / ImageNet | PyTorch |
| deepspeed_cifar10_cpu_offloading | CIFAR-10 | PyTorch (DeepSpeed) |
Natural Language Processing (NLP)
| Example | Dataset | Framework |
|---|---|---|
| albert_squad_pytorch | SQuAD | PyTorch |
| bert_glue_pytorch | GLUE | PyTorch |
| word_language_model | WikiText-2 | PyTorch |
HP Search Benchmarks
| Example | Dataset | Framework |
|---|---|---|
| darts_cifar10_pytorch | CIFAR-10 | PyTorch |
| darts_penntreebank_pytorch | Penn Treebank Dataset | PyTorch |
Neural Architecture Search (NAS)
| Example | Dataset | Framework |
|---|---|---|
| gaea_pytorch | DARTS | PyTorch |
Meta Learning
| Example | Dataset | Framework |
|---|---|---|
| protonet_omniglot_pytorch | Omniglot | PyTorch |
Generative Adversarial Networks (GAN)
| Example | Dataset | Framework |
|---|---|---|
| dc_gan_tf_keras | MNIST | TensorFlow (tf.keras) |
| gan_mnist_pytorch | MNIST | PyTorch |
| deepspeed_dcgan | MNIST / CIFAR-10 | PyTorch (DeepSpeed) |
| pix2pix_tf_keras | pix2pix | TensorFlow (tf.keras) |
| cyclegan | monet2photo | PyTorch |
Custom Reducers
| Example | Dataset | Framework |
|---|---|---|
| custom_reducers_mnist_pytorch | MNIST | PyTorch |
HP Search Constraints
| Example | Dataset | Framework |
|---|---|---|
| hp_constraints_mnist_pytorch | MNIST | PyTorch |
Custom Search Method
| Example | Dataset | Framework |
|---|---|---|
| asha_search_method | MNIST | PyTorch |
Fully Sharded Data Parallel
| Example | Framework |
|---|---|
| minimal_fsdp | PyTorch |