kubeflow
/
examples
mirror of https://github.com/kubeflow/examples.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #508 from CraigSterrett/pytorch_mnist 

Fixed typo in README and one bad link
This commit is contained in:
Michelle Casbon 2019-02-17 09:06:50 -08:00 committed by GitHub
parent 692c78550e 838ad79898
commit ed82693304
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
pytorch_mnist/02_distributed_training.md
View File

@ -39,7 +39,7 @@ kubectl create -f training/ddp/mnist/cpu/v1beta1/job_mnist_DDP_GPU.yaml
With the commands above we have created Custom Resource that has been defined and enabled during Kubeflow
installation, namely `PyTorchJob`.
If you look at [job_mnist_DDP_GPU.yaml](https://github.com/kubeflow/examples/blob/master/github_issue_summarization/training/ddp/mnist/cpu/v1beta1/job_mnist_DDP_GPU.yaml) few things are worth mentioning.
If you look at [job_mnist_DDP_GPU.yaml](https://github.com/kubeflow/examples/blob/master/pytorch_mnist/training/ddp/mnist/gpu/v1beta1/job_mnist_DDP_GPU.yaml) few things are worth mentioning.
1. We mount our shared persistent disk as /mnt/kubeflow-gcfs for our PyTorchJob, where models will be saved at the end of the epochs
2. Allocate one GPU to our container

4
pytorch_mnist/README.md
View File

@ -32,9 +32,9 @@ In other words, you send an image to the model, and the model does its best to i
![](https://www.kubeflow.org/docs/images/gcp-e2e-ui-prediction.png)
In the above screenshot, the image shows a hand-written 8.
In the above screenshot, the image shows a hand-written 7.
The table below the image shows a bar graph for each classification label from 0 to 9. Each bar represents the probability that the image matches the respective label.
Looks like it’s pretty confident this one is an 8!
Looks like it’s pretty confident this one is an 7!
## Steps: