tensorboard-visualizer

Tensorboard Visualizer - Auto-activating skill for ML Training. Triggers on: tensorboard visualizer, tensorboard visualizer Part of the ML Training skill category.

1.00x

Quality

Does it follow best practices?

Impact

100%

1.00x

Average score across 3 eval scenarios

Securityby

Passed

No known issues

Optimize this skill with Tessl

npx tessl skill review --optimize ./planned-skills/generated/07-ml-training/tensorboard-visualizer/SKILL.md

Evaluation results

100%

Diagnosing Model Overfitting with Training Visualization

PyTorch TensorBoard integration

Criteria

Without context

With context

TensorBoard import

100%

SummaryWriter instantiated

100%

Training loss logged

100%

Validation loss logged

100%

Accuracy logged

100%

Logs directory named 'runs/'

100%

Writer closed

100%

TensorBoard in requirements

100%

TensorBoard launch command

100%

PyTorch framework used

100%

Global step passed to add_scalar

100%

Without context: $0.6161 · 2s · 1 turns · 3 in / 33 out tokens

With context: $0.5639 · 6m 21s · 32 turns · 64 in / 6,613 out tokens

100%

Comparing Model Configurations to Find the Best Hyperparameters

Hyperparameter tracking with TensorBoard

Criteria

Without context

With context

TensorBoard used

100%

SummaryWriter per run

100%

add_hparams called

100%

Learning rate as hparam

100%

Metric results in add_hparams

100%

Separate run directories

100%

Multiple configurations

100%

Scalar metrics logged

100%

TensorBoard in requirements

100%

Writers closed

100%

README references HParams or comparison

100%

Without context: $0.8785 · 2s · 1 turns · 3 in / 38 out tokens

With context: $0.6342 · 2s · 1 turns · 3 in / 25 out tokens

100%

Monitoring a Keras Image Classifier During Training

TensorFlow Keras TensorBoard callback

Criteria

Without context

With context

TensorBoard callback used

100%

Callback passed to fit()

100%

Log directory set

100%

Histogram freq enabled

100%

TensorFlow framework used

100%

Keras model defined

100%

Model compiled

100%

TensorBoard in requirements

100%

Validation data provided

100%

README launch command

100%

No manual add_scalar calls needed

100%

Without context: $0.4750 · 2m 18s · 27 turns · 27 in / 5,378 out tokens

With context: $0.5911 · 2m 51s · 36 turns · 330 in / 5,955 out tokens

Repository: jeremylongshore/claude-code-plugins-plus-skills
Commit: f17dd51

Evaluated: 18 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Diagnosing Model Overfitting with Training Visualization Comparing Model Configurations to Find the Best Hyperparameters Monitoring a Keras Image Classifier During Training

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.