Sia HackewrNoon

Table of Links

Preliminaries and Related Work
Key Bottlenecks in PC Parallelization
Harnessing Block-Based PC Parallelization

4.1. Fully Connected Sum Layers

4.2. Generalizing To Practical Sum Layers

4.3. Efficient Implementations by Compiling PC Layers

4.4. Analysis: IO and Computation Overhead
Optimizing Backpropagation with PC Flows
Experiments

6.1. Faster Models with PyJuice

6.2. Better PCs At Scale

6.3. Benchmarking Existing PCs
Conclusion, Acknowledgements, Impact Statement, and References

D.1. Speed of the Compilation Process

In Table 5, we show the compilation speed of PCs with different structures and different sizes. Experiments are conducted on a server with an AMD EPYC 7763 64-Core Processor and 8 RTX 4090 GPUs (we only use one GPU). The results demonstrate the efficiency of the compilation process, where even the PD model with close to 1B parameters can be compiled in around 30 seconds.

Table 5. Average (± standard deviation of 3 runs) runtime (in seconds) of the compilation process of four PCs.

D.2. Runtime on Different GPUs

In addition to the RTX 4090 GPU adopted in the experiments in Table 1, we compare the runtime of PyJuice with the baselines on an NVIDIA A40 GPU. As shown in the following table, PyJuice is still significantly faster than all baselines for PCs of different sizes.

Table 6. Average (± standard deviation of 5 runs) runtime (in seconds) per training epoch of 60K samples for PyJuice and the baselines on five RAT-SPNs (Peharz et al., 2020b) with different sizes. All other settings are the same as described in Section 6.1.

D.3. Runtime on Different Batch Sizes

As a supplement to Table 1, we report the runtime for a RAT-SPN (Peharz et al., 2020b) with 465K nodes and 33.4M edges using batch sizes {8, 16, 32, 64, 128, 256, 512}. To minimize distractions, we only record the time to compute the forward and backward process, but not the time used for EM updates. Results are shown in the table below.

Table 7. Average (± standard deviation of 5 runs) runtime (in seconds) per training epoch (excluding EM updates) of 60K samples for PyJuice and the baselines on a RAT-SPNs (Peharz et al., 2020b) with 465K nodes and 33.4M edges. All other settings are the same as described in Section 6.1. OOM denotes out-of-memory.

Authors:

(1) Anji Liu, Department of Computer Science, University of California, Los Angeles, USA (liuanji@cs.ucla.edu);

(2) Kareem Ahmed, Department of Computer Science, University of California, Los Angeles, USA;

(3) Guy Van den Broeck, Department of Computer Science, University of California, Los Angeles, USA;

This paper is available on arxiv under CC BY 4.0 DEED license.

How Fast Is PyJuice? Testing Compilation Speed Across GPUs and Batch Sizes

Table of Links

D. Additional Experiments

D.1. Speed of the Compilation Process

D.2. Runtime on Different GPUs

D.3. Runtime on Different Batch Sizes