You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I have searched the HUB issues and discussions and found no similar questions.
Question
Hey,
I trained a custom model using the Ultalytics Hub. Now I would like to export the training results (charts, weights, etc). Similar to "runs/detect/train/" when training locally.
Is there such an export feature?
Environment Details
Ultralytics HUB Version: v0.1.40
Client User Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/17.2.1 Safari/605.1.15
Operating System: MacIntel
Browser Window Size: 1920 x 1000
Server Timestamp: 1713615377
The text was updated successfully, but these errors were encountered:
Yes, we support exporting training results like charts and weights from the Ultralytics Hub using our HUB SDK. You can easily accomplish this with the following Python code snippet:
fromhub_sdkimportHUBClient# Initialize the HUBClient with your API keycredentials= {"api_key": "<API-KEY>"}
client=HUBClient(credentials)
print("Authenticated:", client.authenticated)
# Access your trained model by its IDmodel=client.model("<MODEL-ID>")
# Retrieve metrics from the modelmetrics=model.get_metrics()
print("Metrics Data:", metrics)
# Get a URL to download the best model weightsweight_url=model.get_weights_url("best")
print("Weight URL link:", weight_url)
Make sure to replace <API-KEY> and <MODEL-ID> with your actual API key and model ID. This script will authenticate you, fetch the training metrics, and provide a direct link to download the best weights.
Search before asking
Question
Hey,
I trained a custom model using the Ultalytics Hub. Now I would like to export the training results (charts, weights, etc). Similar to "runs/detect/train/" when training locally.
Is there such an export feature?
Environment Details
The text was updated successfully, but these errors were encountered: