Dear organizers,
Would you please provide a step-by-step instruction for participants to submit top 3 models for Midpoint Benchmark Submission?
I could only see the information to acquire the dictionary of cohort study data (sdy1760) as follow instead of submission instruction :
```
The Midpoint Benchmark Submission Deadline is June 25th 2025
The benchmark data dictionary is now available in the cStructure platform (sdy1760)
Open the Data sidebar
Turn the Dictionary switch on and the sdy1760 data dictionary should automatically download*
If the sdy1760 data dictionary does not automatically download:
Delete the sdy1662datadictionary.csv file using the trash can icon in the Data Sources component, close and reopen the Data Sidebar, and switch Dictionary on.
```
Yours,
Tsai-Min
Created by TSAI-MIN (在民) CHEN (陳) chentsaimin Hi Tsai-Min,
Thank you for reporting this issue, we found the problem and pushed a fix. However, we want to extensively test the updated code which will take ~2 days. You can resubmit your models now or wait until the updates have been tested on Tuesday.
We apologize for the inconvenience. Dear organizers,
I tried to submit the same causal model, modelS, for sdy1662 again using the name, modelS0629.
A new error popped up as below, which I never encountered before:
```
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
/app/venv/lib/python3.12/site-packages/pygformula/parametric_gformula/histories.py:86: SettingWithCopyWarning:
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead
See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
pool[lagged_cov] = np.where(pool[time_name] >= lagged_nums[i],
Error during g-formula execution: "['lag1_AGE'] not in index"
```
Could you please help me check it?
Yours,
Tsai-Min Hi @scottie
Do you mean that
```
“sdy1662_modelS” could converge in sdy1662, but “modelS” could not converge in sdy1662?
```
I am baffled by seeing both result page successfully after g-simulation but only
```
“sdy1662_modelS”
```
scored to the leaderboard.
Yours,
Tsai-Min Hello,
I see the following entry in the Leaderboard:
| 27 Metformin-121 sdy1662_modelS 06/20/2025 09:36 AM 714 16 198 500
There is another modelS submission that failed to converge.
I hope that answers your question.
Hi Tsai-Min,
I see the results file which posted on June 20th. I will investigate why the leaderboard was not updated.
Thank you for your patience. Dear organizers,
I have tried to submit my causal model, modelS, for sdy1662 many times.
Althoug I could run the g-formula successfully, I still can't see its score in either 10 or 500 bootstrap leaderboard.
Could you please help me check it?
Yours,
Tsai-Min
Hi Tsai-Min,
The mid-point evaluation leaderboard will be released after June 25th. Each team will be able to submit three models before the final submission date (July 30th) and the mid-point evaluation leaderboard will be updated once the model has been scored.
I hope that answers your question. Dear organizers,
Thanks for your clarification.
We observed there are different features across 2 datasets, sdy1662 and sdy1760.
Thus, we only kept the intersected features from the causal graph designed from sdy1662 training set to submit the reduced causal graph for sdy1760 validation set.
However, we don't know if this reduced graph could be scored successfully in sdy1760 mid-term validation.
Could you please provide us a page, where we could confirm our 3 models scored successfully with corresponding dates and names?
Yours,
Tsai-Min Note: Always use the 'Data Dictionary Source' dropdown to layer features onto your nodes (the FeatureSource is saved as sdy1760.csv). Using 'Node Feature Sources' will not load the available features for sdy1760.csv.
Hello,
We have created a guided tutorial describing the mid-point benchmark submission process. In the cstructure.net project dashboard, click the Tutorials button in the upper left corner of the page and select the Mid-point Benchmark Submission project.
For clarity here are the steps:
**Data Dictionary Download**
Open the Data sidebar
1. Click the Dictionary switch to the On position
2. Click the Data Sources refresh button to confirm the sdy1760_data_dictionary.csv file was downloaded
If the sdy1760 data dictionary was not downloaded:
1. Delete the sdy1662_data_dictionary.csv file using the trash can icon in the Data Sources component
2. Close Data Sidebar
3. Reopen the Data Sidebar
4. Switch Dictionary on
**Data Layering with sdy1760**
Open the Data sidebar
1. Switch Explorer, Layering, and Dictionary components On
2. Ensure the sdy1760_data_dictionary_csv file is Loaded using the switch in the Data Sources component
3. In the Data Layering Component
- Select the sdy1760_data_dictionary_csv file from the Data Dictionary Source dropdown menu
- Layer the sdy1760 data features onto the corresponding node using the Data Layering table interface
Every adjusted node, outcome node, and action node should have a Feature and Datatype associated. This can be confirmed by double-clicking a node in the canvas, clicking the blue 'D' button, and selecting the sdy1760.csv FeatureSource.
**Submitting a Mid-point Benchmark Model**
Open the Estimate sidebar
1. Enter a name for your model
2. Assign nodes to Baseline Covariates or Time-Varying Covariates (STEROIDS should be Time-Varying)
3. Click the orange 'Generate R Script' button at the bottom of the component
4. Click the blue 'Run G-Formula Estimation' button
5. Look for the white 'Benchmark Submitted' pop-up in the lower right corner of the canvas to confirm submission
Drop files to upload
(solved)[Submission] Asking for instruction to Midpoint Benchmark Submission page is loading…