Skip to content

In the project "lung_nodule_ct_detection", how to add the configuration for resuming training and loading a pre-trained model.pt? #1853

In the project "lung_nodule_ct_detection", how to add the configuration for resuming training and loading a pre-trained model.pt?

In the project "lung_nodule_ct_detection", how to add the configuration for resuming training and loading a pre-trained model.pt? #1853

Workflow file for this run

name: chatops
# currently dispatches /black command to project-monai/monai-code-formatter
on:
issue_comment:
types: [created, edited]
jobs:
dispatch_command:
runs-on: ubuntu-latest
steps:
- name: dispatch
uses: peter-evans/slash-command-dispatch@v1.2.0
with:
token: ${{ secrets.PR_MAINTAIN }}
reaction-token: ${{ secrets.GITHUB_TOKEN }}
reactions: false
config: >
[
{
"command": "black",
"permission": "none",
"issue_type": "pull-request",
"allow_edits": true,
"repository": "project-monai/monai-code-formatter"
}
]