Monitoring deployed models in TensorFlow - ML Experiment: Train & Evaluate

Experiment - Monitoring deployed models

Problem:You have deployed a TensorFlow model for image classification. The model performs well initially, but over time, the accuracy on new incoming data drops. You want to monitor the model's performance in real-time to detect when it starts to degrade.

Current Metrics:Initial deployment accuracy: 92%, Current accuracy on new data: 75%

Issue:The model's accuracy decreases over time due to data drift, but there is no monitoring system to alert or log this degradation.

Your Task

Implement a monitoring system that tracks the model's accuracy on new data batches in real-time and logs the metrics. The system should alert if accuracy drops below 85%.

Use TensorFlow for model inference.

Simulate incoming data batches for monitoring.

Do not retrain the model in this experiment.

Hint 1

Hint 2

Hint 3

Solution

TensorFlow

import tensorflow as tf
import numpy as np
from sklearn.metrics import accuracy_score

# Load a pre-trained model (for example purposes, use a simple model)
model = tf.keras.Sequential([
    tf.keras.layers.Flatten(input_shape=(28, 28)),
    tf.keras.layers.Dense(128, activation='relu'),
    tf.keras.layers.Dense(10, activation='softmax')
])

# Compile and train on MNIST for demonstration
(x_train, y_train), (x_test, y_test) = tf.keras.datasets.mnist.load_data()
x_train, x_test = x_train / 255.0, x_test / 255.0

model.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])
model.fit(x_train, y_train, epochs=1, batch_size=64, verbose=0)

# Simulate streaming batches from test data
batch_size = 100
num_batches = len(x_test) // batch_size

for i in range(num_batches):
    batch_x = x_test[i*batch_size:(i+1)*batch_size]
    batch_y = y_test[i*batch_size:(i+1)*batch_size]
    predictions = model.predict(batch_x, verbose=0)
    pred_labels = np.argmax(predictions, axis=1)
    acc = accuracy_score(batch_y, pred_labels) * 100
    print(f"Batch {i+1}/{num_batches} - Accuracy: {acc:.2f}%")
    if acc < 85:
        print(f"ALERT: Accuracy dropped below threshold! Current accuracy: {acc:.2f}%")

Added a loop to simulate real-time incoming data batches.

Calculated accuracy for each batch using sklearn's accuracy_score.

Added print alerts when accuracy falls below 85%.

Results Interpretation

Before monitoring, the model's overall accuracy was 92% initially but dropped to 75% on new data without detection.

After implementing monitoring, batch-wise accuracy is tracked and alerts are raised when accuracy falls below 85%, enabling timely detection of model degradation.

Monitoring deployed models with real-time accuracy tracking helps detect performance drops early, allowing for timely intervention before major failures.

Bonus Experiment

Extend the monitoring system to log accuracy metrics to a file and visualize accuracy trends over time.

💡 Hint

Use Python's logging module to save metrics and matplotlib to plot accuracy after processing multiple batches.

Practice

1. What is the main purpose of monitoring deployed TensorFlow models?

easy

A. To reduce the model size

B. To check model health and accuracy over time

C. To train the model faster

D. To change the model architecture automatically

Solution

1. Step 1: Understand model deployment goals

   After deployment, models must be checked regularly to ensure they perform well and remain accurate.

2. Step 2: Identify monitoring purpose

   Monitoring focuses on tracking model health and accuracy, not training speed or size reduction.

3. Final Answer:

   To check model health and accuracy over time -> Option B

4. Quick Check:

   Monitoring = model health and accuracy check [OK]

Hint: Monitoring means checking model accuracy and health regularly [OK]

Common Mistakes:

• Confusing monitoring with training or model design
• Thinking monitoring changes model architecture
• Assuming monitoring reduces model size

2. Which TensorFlow tool is commonly used to define metrics and slices for monitoring deployed models?

easy

A. TensorFlow Model Analysis (tfma)

B. TensorFlow Lite

C. TensorBoard

D. TensorFlow Hub

Solution

1. Step 1: Identify tools for model monitoring

   TensorFlow Model Analysis (tfma) is designed to evaluate and monitor models by defining metrics and slices.

2. Step 2: Differentiate from other tools

   TensorBoard is for visualization, TensorFlow Lite is for mobile deployment, and TensorFlow Hub is for reusable model components.

3. Final Answer:

   TensorFlow Model Analysis (tfma) -> Option A

4. Quick Check:

   tfma = metrics and slices tool [OK]

Hint: tfma is for monitoring metrics and slices [OK]

Common Mistakes:

• Confusing TensorBoard with tfma
• Using TensorFlow Lite for monitoring
• Mixing TensorFlow Hub with monitoring tools

3. Given the code snippet below, what will be the output of the evaluation metrics?

import tensorflow_model_analysis as tfma

config = tfma.EvalConfig(
  model_specs=[tfma.ModelSpec(label_key='label')],
  metrics_specs=[tfma.MetricsSpec(metrics=[tfma.MetricConfig(class_name='ExampleCount')])]
)

result = tfma.run_model_analysis(
  eval_shared_model=tfma.default_eval_shared_model('model_dir'),
  eval_config=config,
  eval_input_receiver_fn=None
)

print(result.metrics['']['']['ExampleCount'])

medium

A. Error due to missing input_receiver_fn

B. Accuracy of the model

C. Loss value of the model

D. Total number of examples evaluated

Solution

1. Step 1: Analyze the code snippet

   The code calls tfma.run_model_analysis with eval_input_receiver_fn set to None, which is required for input data processing.

2. Step 2: Understand the impact of missing eval_input_receiver_fn

   Without a valid eval_input_receiver_fn, the evaluation cannot run properly and will raise an error.

3. Final Answer:

   Error due to missing input_receiver_fn -> Option A

4. Quick Check:

   Missing input_receiver_fn causes error [OK]

Hint: eval_input_receiver_fn cannot be None for evaluation [OK]

Common Mistakes:

• Assuming metrics print correctly without input_receiver_fn
• Confusing ExampleCount with accuracy or loss
• Ignoring required input functions in tfma

4. You run a TensorFlow Model Analysis job but notice the metrics are not updating after deployment. Which of the following is the most likely cause?

medium

A. The model was trained with too few epochs

B. The training data was too large

C. The model architecture changed after deployment

D. The evaluation job is not scheduled to run regularly

Solution

1. Step 1: Identify why metrics don't update

   If metrics don't update, it usually means the evaluation process is not running after deployment.

2. Step 2: Check scheduling of evaluation jobs

   Evaluation jobs must be scheduled regularly to monitor model performance over time.

3. Final Answer:

   The evaluation job is not scheduled to run regularly -> Option D

4. Quick Check:

   Regular evaluation runs update metrics [OK]

Hint: Schedule evaluation jobs regularly to update metrics [OK]

Common Mistakes:

• Blaming training epochs for monitoring issues
• Assuming architecture changes affect monitoring directly
• Thinking data size stops metric updates

5. You want to monitor a deployed TensorFlow model's performance on different user groups. Which approach best helps you achieve this using TensorFlow Model Analysis?

hard

A. Deploy multiple models without monitoring

B. Retrain the model separately for each user group

C. Define slices in tfma.EvalConfig to segment metrics by user group

D. Use TensorBoard to visualize overall accuracy only

Solution

1. Step 1: Understand slicing in tfma

   Slices allow monitoring metrics for specific data segments, such as user groups, to detect performance differences.

2. Step 2: Apply slices in EvalConfig

   By defining slices in tfma.EvalConfig, you can track metrics separately for each user group without retraining multiple models.

3. Final Answer:

   Define slices in tfma.EvalConfig to segment metrics by user group -> Option C

4. Quick Check:

   Slices segment metrics for groups [OK]

Hint: Use slices in tfma to monitor groups separately [OK]

Common Mistakes:

• Thinking retraining is needed for monitoring groups
• Using TensorBoard only for overall metrics
• Ignoring monitoring and deploying multiple models blindly