Skip to content

Base Model

tensoris.models.model

Base Neural Network Architecture Module.

Classes

BaseModel

Abstract base class for all PyTorch neural network models in the package.

Architecture & Forward Pass Execution Flowchart:

graph TD
    A["Input Data Batch (Images / Tokens / Embeddings)"] --> B["BaseModel.forward(inputs)"]
    B --> C["Feature Extractor / Backbone"]
    C --> D["Neural Layer Blocks (Conv / Attention / Residual)"]
    D --> E["Task Prediction Head (Logits)"]
    E --> F["Loss Function (Focal / Dice / CrossEntropy)"]
    F --> G["Backward Pass & Gradient Calculation"]
    G --> H["Optimizer Step (AdamW / SGD)"]
    H --> I["Metric Logging & Model Checkpoint"]
Source code in src/tensoris/models/model.py 
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
class BaseModel:
    """Abstract base class for all PyTorch neural network models in the package.

    Architecture & Forward Pass Execution Flowchart:

    ```mermaid
    graph TD
        A["Input Data Batch (Images / Tokens / Embeddings)"] --> B["BaseModel.forward(inputs)"]
        B --> C["Feature Extractor / Backbone"]
        C --> D["Neural Layer Blocks (Conv / Attention / Residual)"]
        D --> E["Task Prediction Head (Logits)"]
        E --> F["Loss Function (Focal / Dice / CrossEntropy)"]
        F --> G["Backward Pass & Gradient Calculation"]
        G --> H["Optimizer Step (AdamW / SGD)"]
        H --> I["Metric Logging & Model Checkpoint"]
    ```
    """

    def __init__(self) -> None:
        """Initialize base model hyperparameters."""
        pass

    def forward(self, inputs: list[float]) -> list[float]:
        """Execute a forward pass computation.

        Args:
            inputs: Input tensor data sample.

        Returns:
            Output forward prediction tensor.
        """
        return inputs
Methods:
__init__ 
__init__()

Initialize base model hyperparameters.

Source code in src/tensoris/models/model.py 
22
23
24
def __init__(self) -> None:
    """Initialize base model hyperparameters."""
    pass
forward 
forward(inputs)

Execute a forward pass computation.

Parameters:

Name Type Description Default
inputs list[float]

Input tensor data sample.

 required

Returns:

Type Description
list[float]

Output forward prediction tensor.
Source code in src/tensoris/models/model.py 
26
27
28
29
30
31
32
33
34
35
def forward(self, inputs: list[float]) -> list[float]:
    """Execute a forward pass computation.

    Args:
        inputs: Input tensor data sample.

    Returns:
        Output forward prediction tensor.
    """
    return inputs