Multi-GPU training basics in TensorFlow - Model Metrics & Evaluation

Metrics & Evaluation - Multi-GPU training basics

Which metric matters for Multi-GPU training and WHY

When training a model using multiple GPUs, the key metric to watch is training speed (how fast the model learns). This is often measured by time per epoch or steps per second. Faster training means better use of multiple GPUs.

Besides speed, model accuracy or loss on validation data is important to ensure that using multiple GPUs does not harm the model's learning quality.

We want to confirm that the model trained on multiple GPUs achieves similar or better accuracy compared to single GPU training, while reducing training time.

Confusion matrix or equivalent visualization

Multi-GPU training itself does not change the confusion matrix of the model's predictions. However, here is an example confusion matrix from a classification model trained on multiple GPUs:

      | Predicted Positive | Predicted Negative |
      |--------------------|--------------------|
      | True Positive (TP): 90 | False Negative (FN): 10 |
      | False Positive (FP): 5 | True Negative (TN): 95 |
    

The confusion matrix shows the model's prediction quality. Multi-GPU training aims to keep this quality consistent while speeding up training.

Precision vs Recall tradeoff with concrete examples

Multi-GPU training does not directly affect precision or recall, but it can speed up experiments to find the best balance.

For example, if you train a cancer detection model, you want high recall (catch all cancer cases). Multi-GPU training helps you try different settings faster to improve recall.

Similarly, for spam detection, you want high precision (avoid marking good emails as spam). Faster training lets you test more ideas quickly.

What "good" vs "bad" metric values look like for Multi-GPU training

Good:

• Training time per epoch decreases significantly compared to single GPU (e.g., from 10 minutes to 3 minutes).
• Validation accuracy or loss remains stable or improves.
• No large differences in model predictions or confusion matrix compared to single GPU.

Bad:

• Training speed does not improve or is slower due to overhead.
• Validation accuracy drops significantly, indicating training issues.
• Model predictions become inconsistent or unstable.

Metrics pitfalls in Multi-GPU training

• Data leakage: Not properly shuffling or splitting data across GPUs can cause the same data to be seen multiple times, inflating accuracy.
• Overfitting: Faster training might lead to overfitting if early stopping or validation checks are ignored.
• Accuracy paradox: Speeding up training but ignoring accuracy drop can give a false sense of improvement.
• Synchronization issues: Incorrect gradient averaging across GPUs can harm model convergence.

Self-check question

Your multi-GPU model trains 3 times faster than single GPU but validation accuracy dropped from 90% to 70%. Is this good for production? Why or why not?

Answer: No, it is not good. Although training is faster, the large drop in accuracy means the model learned worse. Speed alone is not enough; model quality must stay high.

Key Result

Multi-GPU training should speed up training time significantly while maintaining or improving model accuracy.

Practice

1. What is the main purpose of using tf.distribute.MirroredStrategy() in TensorFlow?

easy

A. To train a model on multiple GPUs simultaneously

B. To reduce the size of the dataset

C. To convert a model to run on CPU only

D. To save the model in a special format

Solution

1. Step 1: Understand the role of MirroredStrategy

   tf.distribute.MirroredStrategy() is designed to distribute training across multiple GPUs to speed up the process.

2. Step 2: Compare options with the purpose

   Options A, B, and C do not relate to multi-GPU training. Only training a model on multiple GPUs simultaneously correctly states the purpose.

3. Final Answer:

   To train a model on multiple GPUs simultaneously -> Option A

4. Quick Check:

   Multi-GPU training = MirroredStrategy usage [OK]

Hint: MirroredStrategy means training on many GPUs at once [OK]

Common Mistakes:

• Thinking it reduces dataset size
• Confusing with CPU-only training
• Assuming it changes model saving format

2. Which of the following is the correct way to start using multi-GPU training in TensorFlow?

easy

A. model = build_model()
strategy = tf.distribute.MirroredStrategy()
model.compile()

B. strategy = tf.distribute.MirroredStrategy()
with strategy.scope():
model = build_model()

C. strategy = tf.distribute.MirroredStrategy()
model = build_model()
model.fit()

D. model = build_model()
model.fit()
strategy = tf.distribute.MirroredStrategy()

Solution

1. Step 1: Identify correct usage of strategy.scope()

   Model building and compiling must happen inside strategy.scope() to enable multi-GPU support.

2. Step 2: Check each option for correct order

   Only strategy = tf.distribute.MirroredStrategy() with strategy.scope(): model = build_model() builds the model inside the strategy scope. Others build or compile outside, which disables multi-GPU benefits.

3. Final Answer:

   strategy = tf.distribute.MirroredStrategy() with strategy.scope(): model = build_model() -> Option B

4. Quick Check:

   Build model inside strategy.scope() = strategy = tf.distribute.MirroredStrategy() with strategy.scope(): model = build_model() [OK]

Hint: Always build and compile inside strategy.scope() [OK]

Common Mistakes:

• Building model before strategy.scope()
• Compiling model outside strategy.scope()
• Calling fit before defining strategy

3. Given the code below, what will be the output shape of the predictions if the input batch size is 64 and there are 2 GPUs?

strategy = tf.distribute.MirroredStrategy()
with strategy.scope():
    model = tf.keras.Sequential([
        tf.keras.layers.Dense(10, activation='softmax')
    ])
    model.compile(optimizer='adam', loss='sparse_categorical_crossentropy')

import numpy as np
inputs = np.random.random((64, 20))
predictions = model.predict(inputs)
print(predictions.shape)

medium

A. (10, 64)

B. (32, 10)

C. (128, 10)

D. (64, 10)

Solution

1. Step 1: Understand batch size and MirroredStrategy behavior

   MirroredStrategy splits the batch across GPUs internally but the output predictions shape matches the input batch size.

2. Step 2: Check model output shape

   The model outputs 10 classes per input. Input batch size is 64, so output shape is (64, 10).

3. Final Answer:

   (64, 10) -> Option D

4. Quick Check:

   Output shape = (batch_size, classes) = (64, 10) [OK]

Hint: Output shape matches input batch size, not multiplied by GPUs [OK]

Common Mistakes:

• Multiplying batch size by number of GPUs
• Confusing output shape dimensions
• Assuming output shape is (10, 64)

4. You wrote this code but get an error: RuntimeError: tf.distribute.MirroredStrategy() must be used with model creation inside its scope. What is the likely mistake?

strategy = tf.distribute.MirroredStrategy()
model = build_model()
with strategy.scope():
    model.compile(optimizer='adam', loss='mse')

medium

A. Model was compiled outside the strategy.scope()

B. MirroredStrategy cannot be used with compile()

C. Model was built outside the strategy.scope()

D. The optimizer 'adam' is not supported

Solution

1. Step 1: Analyze error message and code

   The error says model creation must be inside strategy.scope(). Here, model is built before entering the scope.

2. Step 2: Identify correct fix

   Building the model inside the scope fixes the error. Compiling inside scope is correct.

3. Final Answer:

   Model was built outside the strategy.scope() -> Option C

4. Quick Check:

   Build model inside scope to avoid RuntimeError [OK]

Hint: Build model inside strategy.scope() to fix this error [OK]

Common Mistakes:

• Building model before strategy.scope()
• Thinking compile() causes error
• Blaming optimizer choice

5. You want to train a large model on 4 GPUs using TensorFlow. Which of the following steps correctly applies multi-GPU training and ensures the model can handle larger batch sizes efficiently?

hard

A. Use tf.distribute.MirroredStrategy(), build and compile the model inside its scope, then increase batch size in model.fit()

B. Build the model first, then use tf.distribute.MirroredStrategy() only during model.fit() with default batch size

C. Use tf.distribute.MirroredStrategy() but keep batch size small to avoid memory issues

D. Train the model separately on each GPU without using MirroredStrategy and average weights manually

Solution

1. Step 1: Apply MirroredStrategy correctly

   Use tf.distribute.MirroredStrategy() and build/compile the model inside its scope to enable multi-GPU training.

2. Step 2: Adjust batch size for efficiency

   Increasing batch size in model.fit() leverages multiple GPUs to handle larger data efficiently.

3. Final Answer:

   Use tf.distribute.MirroredStrategy(), build and compile the model inside its scope, then increase batch size in model.fit() -> Option A

4. Quick Check:

   Build inside scope + increase batch size = efficient multi-GPU training [OK]

Hint: Build inside scope and increase batch size for multi-GPU speed [OK]

Common Mistakes:

• Building model outside strategy scope
• Not increasing batch size to use GPUs fully
• Manually averaging weights instead of using MirroredStrategy