Developer Portal

Appearance

Service Training via SDK ​

Currently, we support training only in the Python SDK, but we are actively working on integrating it into others.

Step 1. Prerequisites ​

This is a quick guide on how to use the training functionality. For a more detailed description of the training and its functionality in the SDK, we recommend reading the following articles:

If you have not yet had experience calling services and using the SDK, we recommend you read the following guides:

Step 2. Install the SDK ​

sh
pip install snet.sdk

Step 3. Configure keys ​

python
import time
from snet import sdk
from snet.sdk.training.responses import ModelStatus


PRIVATE_KEY = "YOUR_PRIVATE_KEY"
ALCHEMY_KEY = "YOUR_ALCHEMY_KEY"

Change private key and alchemy key to your own.

Step 4. Configure variables and base objects ​

python
org_id = "ORG_ID"
service_id = "SERVICE_ID"
group_name = "GROUP_NAME"
grpc_method_name = "METHOD_NAME"

config = sdk.config.Config(private_key=PRIVATE_KEY,
                           eth_rpc_endpoint=f"https://eth-sepolia.g.alchemy.com/v2/{ALCHEMY_KEY}",
                           concurrency=False)

snet_sdk = sdk.SnetSDK(config)
service_client = snet_sdk.create_service_client(org_id=org_id,
                                                service_id=service_id,
                                                group_name=group_name)

You should change the names org_id, service_id, group_name, and grpc_method_name to others depending on which service and method you want to call.

You can also use RPC endpoint of another provider.

Step 5. Create a new model ​

All training functions are called through the training field of the ServiceClient class instance.

Don't forget to save the model id in a variable, it will be useful for further actions.

python
model_name = "YOUR_MODEL_NAME"

new_model = service_client.training.create_model(method_name=grpc_method_name,
                                                 model_name=model_name)
model_id = new_model.model_id

You can also pass other parameters to the create_model method other training methods, more details here.

Step 6. Upload your training data ​

Before uploading the training data, you first need to get the price of uploading and validating the dataset. The training data should be in the form of an archived set of files (e.g. ZIP).

python
validate_price = service_client.training.validate_model_price(model_id)
zip_path = "PATH_TO_YOUR_DATASET_FILE"
model_status = service_client.training.upload_and_validate(model_id, zip_path, validate_price)

After this, you need to wait for the validation to complete. This will be indicated by the model status VALIDATED. Model status can be gotten using the get_model or get_all_models methods.

Step 7. Train your model ​

python
train_price = service_client.training.train_model_price(model_id)
model_status = service_client.training.train_model(model_id, train_price)

Training the model can also take some time, so you need to wait for that too. Completion of training will be indicated by the model status READY_TO_USE.

Step 8. Call the service! ​

Now you can call the method using your new model in the same way you would call a regular service. The only difference is that you also need to pass model_id to the call_rpc method.

python
grpc_message_name = "RPC_MESSAGE_NAME"
parameters = {
    "param1": "value1",
    "param2": "value2"
}

result = service_client.call_rpc(grpc_method_name, grpc_message_name, model_id=model_id, **parameters)

Don't forget to change grpc_message_name and also keys and values of parameters.