Hello ,

I am doing hyperparameter tuning using unified classification. But I am getting optimal parameter as "None" in report. I tried both RandomSearchCV and GridSearchCV. Can you please tell me where I am doing wrong.

rfc = UnifiedClassification(func = 'RandomDecisionTree')

gscv = GridSearchCV(estimator=rfc,
param_grid={'n_estimators': [50,80,90,100,200,300,400,500,600,800],
'max_depth': [50, 60, 70, 80, 90, 100, 110,120,150],
'split_threshold': [0.1, 0.4, 0.7, 1],
'min_samples_leaf':[1,2,3,4]},
train_control=dict(fold_num=5,
random_state=1,
resampling_method='cv'),
scoring='error_rate')

gscv.fit(data=hdf_train, key= 'id',
label='label',
features=features,
partition_method='stratified',
partition_random_state=1,
stratified_column='label',
build_report=True)

Hello ,

while using Grid Search CV with unified classification I am getting below error :

ERROR:hana_ml.algorithms.pal.unified_classification:(423, 'AFL error: "HC_APL"."(DO statement)": line 65 col 1 (at pos 2182): search table error: _SYS_AFL.AFLPAL:UNIFIED_CLASSIFICATION_ANY: [423] (range 3) AFL error exception: exception 73001043: PAL error[73001043]:Value [0] of parameter TRY_NUM is invalid. \n')

I haven't declared any "TRY_NUM " variable in my code. Can you please tell me where I am doing wrong.

code :
`# train model with hyperparameter tunning
rfc = UnifiedClassification(func = 'randomforest')

gscv = GridSearchCV(estimator=rfc,
param_grid={'n_estimators': [50,80,90,100,200,300,400,500,600,800],
'max_depth': [50, 60, 70, 80, 90, 100, 110,120,150],
'split_threshold': [0.1, 0.4, 0.7, 1],
'min_samples_leaf':[1,2,3,4]},
train_control=dict(fold_num=5,
random_state=1,
resampling_method='cv'),
scoring='error_rate')

gscv.fit(data=hdf_train, key= 'CC_KEY', label='NAT_FUNCTION',partition_random_state=1,
features=features,
partition_method='stratified',
stratified_column='class',
build_report=True)
`

A violation against the OSS Rules of Play has been detected.

Rule ID: rl-assigned_teams-3
Explanation: Does it have enough admins on GitHub? No

Find more information at: https://sap.github.io/fosstars-rating-core/oss_rules_of_play_rating.html

The OSPO bot created this issue by mistake - It did not have enough priviledge to check the vulnerability alerts, So I am closing this issue now. Sorry for any inconvenience.

Hello,

I currently have a table to save data, the idea is that new data is constantly being introduced and saved/append to the last rows of said table.
Our team currently generates several indexes on this table:

I'm currently dealing with two issues:

1. I currently have 4000 rows in the table, I'm adding 2000 new rows (with the same columns) via Pandas Dataframes, from Pandas I'm handling this new 6000 rows dataframe (old 4000+new 2000) and pushing it using the following code
   hana_ml.dataframe.create_dataframe_from_pandas(cc, df_upload, 'BUPRY', schema=schema, force=True, primary_key='INDEX')

However with this the table gets dropped and a new one gets created with the new uploaded data, in this process the "INDEXES" also get cleaned up.

2. I have been trying to not drop the table to prevent said "INDEXES" tab from getting cleaned, mainly playing with the two following parameters:

• force=False/True
• drop_exist_tab=False/True

When changing the value of these to both False I have noticed the following warning on the console
Table already exists. Data will be appended to the existing table. To set "force=True" can empty the table.
Which is the correct outcome I am looking for, however I am getting the following error.

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "D:\bupryhr_dev\venv\lib\site-packages\hana_ml\dataframe.py", line 4658, in create_dataframe_from_pandas
    cursor.executemany(sql, rows)
hdbcli.dbapi.IntegrityError: (301, 'unique constraint violated')

I have been trying changing different parameters, uploading the total 6000 rows and just the 2000 new ones, not specifying the primary key, but nothing seems to work and I keep getting the same error.

How can I fix this? Thanks in advance.

Hello ,
Do we need hyperparameter tuning while using SAP APL models ?
as per document there are default values for learning_rate,max depth ,max_iterations,early stopping . After using this default values my model is giving me 81% ACC on training dataset. But I need to further improve the model ACC. I got nowhere about hyper-parameter tuning about APL models.
In APL, how can I tune above parameters , do we have any options to activate correct auto-selection for above hyperparameters or model do it by themselves at backend and no need to do any hyperparameter tuning for APL models ?

my code :
apl_model = GradientBoostingClassifier() apl_model.set_params(variable_auto_selection = True,variable_selection_max_nb_of_final_variables = '6') apl_model.fit(hdf_train, label=col_target, key=col_key,features=col_predictors)

Hi ,
Do we have anything option like in HGBT algorithm to execute parallel run in model training ?
so Parallel runs will complete model training in minimum timestamp

Hello ,
I'm unable to understand below error for hana_ml version --> 2.14.22120100

Error :
ERROR:hana_ml.algorithms.pal.unified_classification:(423, 'AFL error: "HC_APL"."(DO statement)": line 60 col 1 (at pos 1957): search table error: _SYS_AFL.AFLPAL:UNIFIED_CLASSIFICATION_ANY: [423] (range 3) AFL error exception: exception 73001255: PAL error[73001255]:Invalid model. Model format does not match specification.. :nodes/n\n')

I'm using PAL algorithm as below :

from hana_ml.algorithms.pal.unified_classification import UnifiedClassification
rdt_params = dict(random_state=2,n_estimators=10,max_depth=25,learning_rate=0.1)
uc_rdt = UnifiedClassification(func = 'HybridGradientBoostingTree', thread_ratio=1.0,**rdt_params)
uc_rdt.fit(data=res,
              key= 'ID', 
              label='Target',
              features=features,
              partition_method='stratified',
              stratified_column='Target', 
              partition_random_state=2,
              training_percent=0.8, ntiles=2)

I have cross checked already that features are available in dataset as well as named it properly . It works when I ran it yesterday. but today for same dataset and same code --> it's giving me above error.

Can anyone please help me to understand the root cause of this error.

I'm at step 1 of the tutorial: https://developers.sap.com/tutorials/hana-cloud-python-analysis-multimodel-3.html

Error I get is:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
Cell In[20], [line 1](vscode-notebook-cell:?execution_count=20&line=1)
----> [1](vscode-notebook-cell:?execution_count=20&line=1) g_storm.describe()

File [~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:196](https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/gunter/llm_graph/~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:196), in Graph.describe(self)
    [194](https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/gunter/llm_graph/~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:194) # Merge all to one Series
    [195](https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/gunter/llm_graph/~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:195) desc_series = pd.Series(desc)
--> [196](https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/gunter/llm_graph/~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:196) desc_series = pd.concat(desc_series, describer.self_loops)
    [197](https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/gunter/llm_graph/~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:197) desc_series = pd.concat(desc_series, describer.degree)
    [198](https://vscode-remote+wsl-002bubuntu.vscode-resource.vscode-cdn.net/home/gunter/llm_graph/~/llm_graph/.venv/lib/python3.11/site-packages/hana_ml/graph/hana_graph.py:198) desc_series = pd.concat(desc_series, describer.density)

TypeError: concat() takes 1 positional argument but 2 were given

I read that some pandas version (1.5) allowed to have 2 arguments, but not pandas 2.2.2:
https://stackoverflow.com/questions/76034809/pandas-concat-second-argument

Could it be a bug?

A violation against the OSS Rules of Play has been detected.

Rule ID: rl-assigned_teams-1
Explanation: Does it have enough teams on GitHub? No

Find more information at: https://sap.github.io/fosstars-rating-core/oss_rules_of_play_rating.html

ERROR: Could not find a version that satisfies the requirement hdbcli (from versions: none)
ERROR: No matching distribution found for hdbcli

ERROR: Could not find a version that satisfies the requirement hana-ml (from versions: none)
ERROR: No matching distribution found for hana-ml

Hello,

I am looking for GridSearchCV in SAP HANA documents . I am not able to find the best_estimator_ attribute in this library. For example after using GridSearchCV how do I know which attributes are get selected by the program and what is the accuracy for it.

Can you please share documents where I can get idea of it. I already looked at SAP HANA document (not much useful)

Hello ,
I have some doubts regarding model_debriefing in SAP HANA ml package as below :

1. Can we perform model_debriefing on hybrid gradient boosting tree classification APL algorithm? also what about PAL ?
2. My intention is only I want to visualize the model , I mean how it works at backend while doing prediction , is it possible ?
3. I have tried using hana_ml.visualizers.model_debriefing on HGBT classification model (more interested on APL ). But some how it is not working .
   

It will be really helpful if you share your points with us .
Thank you !

A violation against the OSS Rules of Play has been detected.

Rule ID: rl-assigned_teams-2
Explanation: Does it have an admin team on GitHub? No

Find more information at: https://sap.github.io/fosstars-rating-core/oss_rules_of_play_rating.html

Due to the missing Jinja2 dependency, most of the library's features can not be imported. It cause an import error, resulting in a non working version of hana_ml .

These below two munally installation will fix the problem. As this must be done manually - I strongly recommend integrating this dependency into hana_ml pip packages as an dependancy

pip install jinja2
or
add poetry jinja2

Hello ,

RDT Classifier and Random Forest in SAP HANA ML package, are both same algorithms ?
I have checked documentation of SAP , attributes in both algorithms looks like same for me, Can you also please confirm from your end as well .
Currently I am trying to use Grid Search CV with unified classification but there is no option for Random Forest algorithm. Hence Raising this issue.