> ## Documentation Index > Fetch the complete documentation index at: https://arize-ax.mintlify.dev/docs/llms.txt > Use this file to discover all available pages before exploring further. # Pandas SDK Example > Log model inference data directly to Arize using SDK/API methods This page shows how to send data to Arize using the [Python Pandas SDK](/ax/machine-learning/machine-learning/api-reference-ml/python-sdk/arize-pandas). ### Step 1: Set Up Python SDK Install Arize SDK ```bash theme={null} %pip install -q "arize<8.0.0" ``` Initialize Arize client from `arize.pandas.logger` to call `Client.log()` ```python theme={null} from arize.pandas.logger import Client, Schema API_KEY = 'ARIZE_API_KEY' SPACE_ID = 'YOUR SPACE ID' arize_client = Client(space_id=SPACE_ID, api_key=API_KEY) ``` ### Step 2: Set Model Schema Attributes A model schema is broken into required and optional parameters. Optional model schema parameters vary based on model types. Learn more about model types [here](/ax/machine-learning/machine-learning/use-cases-ml). Gain a comprehensive list of schema attributes and their definitions [here](/ax/machine-learning/machine-learning/concepts-ml/model-schema-reference). #### Example Row | prediction\_id | prediction\_ts | prediction\_label | actual\_label | state | states | gender | vector | text | image\_link | | -------------- | -------------- | ----------------- | ------------- | ----- | --------- | ------ | ------------------------ | ------------------------- | -------------------------- | | 1fcd50f4689 | 1637538845 | No Claims | No Claims | ca | \[ca, ak] | female | \[1.27346, -0.2138, ...] | "This is an example text" | "https\://example\_ur.jpg" | ```javascript theme={null} schema = Schema( prediction_id_column_name="prediction_id", timestamp_column_name="prediction_ts", prediction_label_column_name="prediction_label", actual_label_column_name="actual_label", feature_column_names=["state", "states"], tag_column_names=["gender"] ) ``` #### *Optional: Typed Columns* See [Sending Data FAQ](/ax/machine-learning/machine-learning/how-to-ml/upload-data-to-arize/sending-data-faq#how-can-i-ensure-that-consistent-types-are-logged-across-multiple-sdk-calls) for more info on SDK typing features. ```python theme={null} # feature & tag columns can be optionally defined with typing: feature_columns = TypedColumns( inferred=["state"], to_str=["states"], ) schema = Schema( prediction_id_column_name="prediction_id", timestamp_column_name="prediction_ts", prediction_label_column_name="prediction_label", actual_label_column_name="actual_label", feature_column_names=feature_columns, tag_column_names=["gender"] ) ``` #### *Optional: Embeddings* ```python theme={null} # Declare embedding feature columns embedding_feature_column_names = { # Dictionary keys will be the name of the embedding feature in the app "embedding_display_name": EmbeddingColumnNames( vector_column_name="vector", # column containing embedding vector (required) data_column_name="text", # column containing raw text (optional NLP) link_to_data_column_name="image_link" # column containing image URL links (optional CV) ) } schema = Schema( embedding_feature_column_names=embedding_feature_column_names, ... ) ``` #### *Optional: SHAP Values* ```python theme={null} # Generate the Shap Values and save as a Dataframe explainer = shap.TreeExplainer(tree_model) shap_values = explainer.shap_values(X_data) shap_dataframe = pd.DataFrame( shap_values, columns=[f"{fn}_shap" for fn in data["state"]] ) shap_cols = shap_dataframe.columns schema = Schema( shap_values_column_names=dict(zip("state", shap_cols)), ... ) ``` #### *Optional: Delayed Actuals* If your model receives delayed actuals, log your delayed production data using the same **prediction ID**, which links your files together in the Arize platform. This can be delivered days or weeks after the prediction is received. ```python theme={null} #log predictions schema = Schema( prediction_id_column_name="prediction_id", prediction_label_column_name="prediction_label", ... ) # then log actuals schema = Schema( prediction_id_column_name="prediction_id", #needs to be the same as above actual_label_column_name="actual_label", ... ) ``` ### Step 3: Log Inferences Arize expects the DataFrame's index to be sorted and begin at 0. If you perform operations that might affect the index prior to logging data, reset the index as follows: ``` example_dataframe = example_dataframe.reset_index(drop=True) ``` ```python theme={null} response = arize_client.log( model_id='sample-model-1', model_version='v1', #Specify your model version to easily track changes across the Arize platform (i.e. a retrained model) batch_id=None, model_type=ModelTypes.BINARY_CLASSIFICATION, metrics_validation=[Metrics.CLASSIFICATION, Metrics.AUC_LOG_LOSS] environment=Environments.PRODUCTION,#pick from training, production,or validation data dataframe=example_dataframe, schema=schema ) ``` #### *Optional: Metrics Validation* ```swift theme={null} metrics_validation=[Metrics.CLASSIFICATION, Metrics.AUC_LOG_LOSS] ``` There is an optional argument that specifies desired groups of metrics for validation. Combined with a model\_type and based on the schema, Arize will validate that these expected metrics will be available in the platform, and will validate required schema columns. Call \_\_repr\_\_() on a Metrics enum to see its description: ```swift theme={null} repr(Metrics.CLASSIFICATION) > CLASSIFICATION metrics include: Accuracy, Recall, Precision, FPR, FNR, F1, Sensitivity, Specificity ``` Learn more about metrics families [here](/ax/machine-learning/machine-learning/how-to-ml/monitors/choosing-your-metrics#performance-monitors). ### Other Supported SDKs * [Python Pandas SDK](/ax/machine-learning/machine-learning/api-reference-ml/python-sdk/arize-pandas) (log a pandas dataframe) * [Python Single Record SDK](/ax/machine-learning/machine-learning/api-reference-ml/python-sdk/arize-log/client) (log a single record) * [Java SDK](/ax/machine-learning/machine-learning/api-reference-ml/java-sdk) * [R SDK](/ax/machine-learning/machine-learning/api-reference-ml/r-sdk) * [gRPC API](/ax/machine-learning/machine-learning/api-reference-ml/grpc-api) ### Tutorials on how to log predictions, actuals, and feature importance. | | | | ------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------- | | Logging Predictions Only | [Colab Link](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_Pred_Only.ipynb) | | Logging Predictions First, Then Logging Delayed Actuals | [Colab LInk](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_Delayed_Actuals.ipynb) | | Logging Predictions First, Then Logging SHAPs After | [Colab Link](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_Delayed_SHAP.ipynb) | | Logging Predictions and Actuals Together | [Colab Link](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_Pred_Actual_Only.ipynb) | | Logging Predictions and SHAP Together | [Colab Link](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_Pred_SHAP_Only.ipynb) | | Logging Predictions, Actuals, and SHAP Together | [Colab Link](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_Pred_Everything.ipynb) | | Logging PySpark DataFrames | [Colab Link](https://colab.research.google.com/github/Arize-ai/tutorials_python/blob/main/Arize_Tutorials/Log_Examples/Arize_Tutorial_PySpark.ipynb) |