Model training report

{{ exec.end_time }}

Execution details

Algorithm{{ exec.model_name }}
Run ID{{ exec.run_id }}
Duration{{ exec.duration }}
{% if software_section %}

Software

Torch version{{ exec.torch_version }}
Mlektic version{{ exec.mlektic_version }}
Python version{{ exec.python_version }}
{% endif %} {% if hardware_section %}

Hardware / System

Device{{ exec.device }}
GPU{{ exec.gpu_name }}
Total RAM{{ exec.ram_total }}
Host{{ exec.host }}
Operating system{{ exec.os }}
Seed{{ exec.seed }}
{% endif %}

Dataset

{% if dataset.preprocessing_text %} {% elif dataset.preprocessing_list %} {% elif dataset.preprocessing_note %} {% endif %}
Source{{ dataset.name }}
Dimensions{{ dataset.dims }}
Target variable{{ dataset.target }}
Input features (≤50){{ dataset.features_show }}
Data split Train: {{ dataset.split.train.n }} ({{ '%.1f'|format(dataset.split.train.pct) }}%) | Validation: {{ dataset.split.val.n }} ({{ '%.1f'|format(dataset.split.val.pct) }}%) | Test: {{ dataset.split.test.n }} ({{ '%.1f'|format(dataset.split.test.pct) }}%)
Preprocessing steps {{ dataset.preprocessing_text }}
Preprocessing steps
    {% for step in dataset.preprocessing_list %}
  • {{ step }}
    • {% endfor %}
Preprocessing (note) {{ dataset.preprocessing_note }}
{% if dataset.classes %}

Class distribution

{% for c,v in dataset.classes.items() %} {% endfor %}
ClassFrequency
{{ c }}{{ v }}
{% if figs.class_pie %}
{% endif %} {% endif %}

Hyperparameters & settings

{% for k,v in hparams.items() %} {% endfor %}
{{ k }}{{ v|safe }}
{% if results.metrics_table %}

Results

{% for row in results.metrics_table %} {% endfor %}
MetricTrainValidationTest
{{ row.name }} {{ row.train }} {{ row.val }} {{ row.test }}
{% endif %} {% if results.confusion_matrix %}

Confusion matrix

{{ results.confusion_matrix | safe }} {% endif %} {% if figs.loss_curve or figs.metric_curve %}

Training evolution

{% if figs.loss_curve %} {% endif %} {% if figs.metric_curve %} {% endif %} {% endif %} {% if params_sample %}

Parameter sampling during training

{% for p in params_sample %} {% endfor %}
ParameterStartMidEnd
{{ p[0] | safe }} {{ p[1] | safe }} {{ p[2] | safe }} {{ p[3] | safe }}
{% endif %} {% if equation %}

Learned model

{{ equation | safe }}
{% endif %}

Report generated with mlektic: /github.com/DanielDialektico/mlektic