Initialize the graph learning model. Args: input_dim (int): Input feature dimension hidden_dim (int): Hidden

Decoder network for VAE. Args: out_channels (int): Number of output channels latent_dim (int): Dimension of l

VAE with Finite Scalar Quantization. Args: in_channels (int): Number of input channels latent_dim (int): Dime

Initialize Finite Scalar Quantization module. Args: num_levels (int): Number of quantization levels embedding_dim (in

Initialize trainer. Args: model: Model to train train_loader: Training data loader test_loader: T

CIFAR-10 dataset with preprocessing. Args: root_dir (str): Directory containing dataset train (bool): If True

Initialize Inception Score calculator. Args: device: Device to run calculations on

Initialize evaluator. Args: model: Model to evaluate test_loader: Test data loader device: Device

Initialize VQ-VAE model evaluator Args: model (nn.Module): Trained VQ-VAE model test_loader (DataLoa

log_path: path of log file, None

Initialize the GitHub client Args: token: GitHub Personal Access Token, if None, try to get from environment var

Initialize the GitHub searcher Args: token: GitHub Personal Access Token, optional

Memory: memory and external knowledge management. Args: project_path: the path to store the data. embedding_m

Perform a Bing web search, and return the results formatted in Markdown. Args: bing_api_key: key for the Bing search API

Instantiate a new RequestsMarkdownBrowser. Arguments: start_page: The page on which the browser starts (default: "about:

Clicks the mouse on the target with the given element bid. Examples: _click_id('12') _click_id('12', button='left')

Get the markdown content of the current page Examples: _get_page_markdown()

获取单个文件的内容

Convert a local path to a docker path local_path: the local path to convert, like `{local_workplace}/downloads/xxx`

Visit a page, including downloading files based on the url Examples: _visit_page('https://archive.org/download/higpt_stage2/instruct

Return the address of the current page.

Run an agent with a given model, agent function, query, and context variables. Args: model (str): The name of the model. agen

Display a question in a panel and prompt the user for an answer. :param question: the question to display. :param title: the title of the

Use this function when the case is not resolved when all agents have tried their best. [IMPORTANT] Please do not use this function unless all

The failure reason is the reason why you cannot find a solution to the task. You can use this function only after you have tried multiple times an

The task response is the result of the task. Use this function only after you have successfully completed the task. Args: task_response

Use this function to given the analysis report of exsiting experiments and the further plan to the `Machine Learning Agent` to do more experiment

After you have taken enough notes for the innovation, you should use this function to merge the notes for the further innovation.

The function to output the determined reference codebases. Use this function only after you have carefully reviewed the existing resources and un

Use this function when you have finished the task and want to give a suggestion about the implementation. You can only use this function after yo