TiKit is an open-source SDK for Python provided by TI Platform, which can be used to submit training tasks to TI Platform.
Currently, Python 3.4 or later is supported.
On TI Platform, TiKit is built in the containers for notebooks and training tasks, so you don't need to install it separately.
In a non-public cloud TI Platform environment, the installation methods are as follows:
# CentOS: sudo yum -y install cyrus-sasl cyrus-sasl-devel cyrus-sasl-lib # Ubuntu: sudo apt-get update sudo apt-get install -y libsasl2-dev
pip3 install -U tikit
pip3 install tikit-1.0.0-py3-none-any.whl # Alternatively, after decompressing the source code, run python3 setup.py install
Log in to Tencent Cloud to get them on the corresponding page as shown below:
from tikit.client import Client # Initialize the client. In a public cloud TI Platform environment (including notebooks and training tasks), you can leave the region information empty, which is already contained in the environment variable. client = Client("your_secret_id", "your_secret_key", "<region>") # View the list of algorithm frameworks. client.describe_training_frameworks()
In a notebook environment, the HTML content will be directly displayed as shown below:
In a non-notebook environment, you can print the result as shown below: