Closed lmrios closed 1 year ago
Hi @lmrios
Prince here,
Thanks for getting in touch!
I ran your code using Colab, and it worked without issues.
run['MyProcess/training/demo_number']= 1
run['MyProcess/training/demo_dict']= {'Name': "Tom", 'Age': 20}
demo_data = {'Name': ['Tom', 'nick', 'krish', 'jack'], 'Age': [20, 21, 19, 18]}
demo_data_df = pd.DataFrame(demo_data)
run['MyProcess/training/demo_df'].upload(File.as_html(demo_data_df))
Example run: https://app.neptune.ai/o/common/org/showroom/e/SHOW-30049/all?path=MyProcess%2Ftraining%2F&attribute=demo_df
Can you send me a fully reproducible example including your neptune initialization? Note: Please omit your project and api_token arguments.
Hi @lmrios
Just checking in to see if you still need help with this :)
Hi, I am still cleaning my code. Perhaps something wrong is on my side. Thank you for reaching out. You can close this. Best
After running the train process for may model, I want to save a Pandas DataFrame with some metrics inside a method of the training class. It seems .upload() is not looking in the right path I am running GoogleColab, after the training cell crashed. I wrote the following code. Please see Option1 it is the closest to my code
I managed to solve it but I dont want to save my DF to disk
Environment
The output of
pip list
:Google Colab Python 3.8.10