PyTorch

Learn about PyTorch conda environments.

A description of the PyTorch 2.1 for GPU on Python 3.9 with CUDA 11.8 (version 1.0) conda environment.


Released	February 9, 20204
Description	PyTorch is a machine learning library that's used for applications in computer vision and natural language processing. It provides high level features for tensor computing and deep neural networks. Use oracle-ads to speed up your data science workflow with tools that automate common tasks. This environment includes CUDA 11.8 and libraries commonly used in training large language models (LLMs). To get started with the PyTorch environment, open and review the notebook example getting-started.ipynb, using the Launcher.
Python Version	3.9
Slug	`pytorch21_p39_gpu_v1`
Object Storage Path	`oci://service-conda-packs@id19sfcrra6z/service_pack/gpu/PyTorch_2.1_for_GPU_on_Python_3.9/1.0/pytorch21_p39_gpu_v1`
Top Libraries	oracle-ads (v2.10.0) accelerate (v0.26.1) cuda-toolkit (v11.8.0) deepspeed (v0.12.3) peft (v0.7.1) pytorch (v2.1.2) pytorch-lightning (v2.1.2) tensorboard (v2.12.1) transformers (v4.35.2) torchvision (v0.16.1) langchain (v0.1.1) triton-model-analyzer (v1.33.0) For a complete list of preinstalled Python libraries, see pytorch21_p39_gpu_v1.txt.

A description of the PyTorch 2.0 for GPU on Python 3.8 with CUDA 11.8 (version 2.0) conda environment.


Released	September 12, 2023
Description	PyTorch is a machine learning library that's used for applications in computer vision and natural language processing. It provides high level features for tensor computing and deep neural networks. Use oracle-ads to speed up Data Science workflows with tools that automate common tasks. This environment includes CUDA 11.8 and libraries commonly used in training large language models (LLMs). To get started with this conda environment, review the Getting Started notebook, using the Launcher.
Python Version	3.8
Slug	`pytorch20_p39_gpu_v2`
Object Storage Path	`oci://service-conda-packs@id19sfcrra6z/service_pack/gpu/PyTorch_2.0_for_GPU_on_Python_3.9/2.0/pytorch20_p39_gpu_v2`
Top Libraries	oracle-ads (v2.8.9) accelerate(v0.20.3) cuda-toolkit(v11.8.0) deepspeed (v0.10.0) peft (v0.4.0) pytorch (v2.0.1) pytorch-lightning (v2.0.1) tensorboard (v2.12.1) transformers (v4.24.0) torchvision (v0.15.2) langchain (v0.0.267) For a complete list of preinstalled Python libraries, see pytorch20_p39_gpu_v2.txt.

A description of the PyTorch 2.0 for GPU on Python 3.8 (version 1.0) conda environment.


Released	July 10, 2022
Description	PyTorch is a machine learning library that's mainly used for applications in computer vision and natural language processing It provides high level features for tensor computing and deep neural networks. Use `oracle-ads` to speed up your data science work flow with tools that automate common tasks. To get started with this conda environment, review the Getting Started notebook, using the Launcher.
Python Version	3.8
Slug	`pytorch20_p39_gpu_v1`
Object Storage Path	`oci://service-conda-packs@id19sfcrra6z/service_pack/gpu/PyTorch_2.0_for_GPU_on_Python_3.9/1.0/pytorch20_p39_gpu_v1`
Top Libraries	oracle-ads (v2.8.7) pytorch (v2.0.1) pytorch-lightning (v2.0.1) tensorboard (v2.11.0) transformers (v4.24.0) torchvision (v0.15.2) deepspeed (v0.9.4) accelerate (v0.20.3) For a complete list of preinstalled Python libraries, see pytorch20_p39_gpu_v1.txt.

A description of the PyTorch 1.10 for GPU on Python 3.8 (version 1.0) conda environment.


Released	December 1, 2022
Description	PyTorch is a machine learning library that is mainly used for applications in computer vision and natural language processing It provides high level features for tensor computing and deep neural networks. Use `oracle-ads` to speed up your data science work flow with tools that automate common tasks. To get started with this conda environment, review the Getting Started notebook, using the Launcher.
Python Version	3.8
Slug	`pytorch110_p38_gpu_v1`
Object Storage Path	`oci://service-conda-packs@id19sfcrra6z/service_pack/gpu/PyTorch_1.10_for_GPU_on_Python_3.8/1.0/pytorch110_p38_gpu_v1`
Top Libraries	oracle-ads (v2.6.8) pandas (v1.4.4) scikit-learn (v1.1.2) pytorch (v1.10.2) pytorch-lightning (v1.7.7) tensorboard (v2.10.1) transformers (v4.18.0) torchvision (v0.11.3) For a complete list of preinstalled Python libraries, see pytorch110_p38_gpu_v1.txt.