NCA-GENL Data Analysis and Visualization Practice Question
During a fine-tuning run you observe that the training loss decreases smoothly, but validation loss begins rising after epoch 3. You want a single visualization that makes this divergence and the resulting overfitting point immediately obvious to reviewers. Which plot should you produce?
⚠ Common exam trap
The trap here is reporting summary numbers or distributions that describe loss magnitude while omitting the epoch ordering that actually demonstrates when overfitting began.
Answer choices
Why each option matters
Answer the question above first, then reveal the full breakdown to understand why each option is right or wrong.
Correct answer & explanation
✓
A line chart of training and validation loss versus epoch on the same axes
Overfitting is a temporal phenomenon: the two loss curves move together, then separate. A line chart with epoch on the x-axis and both curves on the y-axis preserves that trajectory and makes the inflection where validation loss turns upward easy to locate. Aggregating to a single epoch, plotting noisy batch points, or collapsing across epochs all remove the temporal evidence needed to identify the divergence.
Answer analysis
Option-by-option breakdown
For each option: why learners choose it and why it is or isn't the right answer here.
- ✓
A line chart of training and validation loss versus epoch on the same axes
Why this is correct
Plotting both loss curves against epoch on shared axes makes the divergence explicit: training loss continues downward while validation loss turns upward after epoch 3. The crossover region where validation loss stops improving is visually unmistakable, giving reviewers a direct, quantitative picture of when overfitting began and how large the gap has grown since.
- ✗
A histogram of validation loss values across all epochs
Why it's wrong here
A histogram collapses the epoch dimension, showing how validation losses are distributed but not when they changed. It cannot reveal that early epochs had lower validation loss and later ones higher, nor can it show the corresponding training trend. The temporal relationship between the two curves, which is the core evidence of overfitting, is entirely lost in this representation.
- ✗
A bar chart comparing final training loss and final validation loss at the last epoch
Why it's wrong here
A bar chart of only the final epoch discards the trajectory that defines overfitting. It cannot show when validation loss began rising or that training loss kept falling, so reviewers cannot see the divergence point. Summarizing two numbers loses the temporal evidence needed to justify early stopping or regularization, making this chart inadequate for the stated purpose.
- ✗
A scatter plot of individual batch losses colored by epoch
Why it's wrong here
Batch-level losses are noisy and overlapping, so the trend is buried in variance. Coloring by epoch adds a dimension but still requires mental aggregation to see the divergence. Without a smoothed per-epoch summary for each split, reviewers cannot easily identify the epoch where validation loss began to rise, defeating the goal of making overfitting obvious at a glance.
About these practice questions
This NCA-GENL question is part of Courseiva's 367-question bank — original exam-style content with full explanations and wrong-answer analysis, never real exam questions or exam dumps. Learn why practice questions differ from exam dumps →
JA
Written and reviewed by Johnson Ajibi, MSc IT Security
Senior Network & Security Engineer · founder of Courseiva
Last reviewed September 2026 · checked against the official NVIDIA exam blueprint
This NCA-GENL practice question is part of Courseiva's free NVIDIA certification practice question bank. Courseiva provides original exam-style practice questions with explanations, topic-based practice, mock exams, readiness tracking, and study analytics to help learners prepare for the NCA-GENL exam.