400Ping commented on PR #928:
URL: https://github.com/apache/mahout/pull/928#issuecomment-3796842232

   > Before:
   > 
   > ```
   > $ python qdp-python/benchmark/benchmark_throughput.py \
   >   --qubits 16 --batches 3200 --batch-size 256 --prefetch 32 \
   >   --frameworks mahout --encoding-method angle
   > Generating 819200 samples of 16 qubits...
   >   Batch size   : 256
   >   Vector length: 65536
   >   Batches      : 3200
   >   Prefetch     : 32
   >   Frameworks   : mahout
   >   Encode method: angle
   >   Generated 819200 samples
   >   PennyLane/Qiskit format: 409600.00 MB
   >   Mahout format: 409600.00 MB
   > 
   > ======================================================================
   > DATALOADER THROUGHPUT BENCHMARK: 16 Qubits, 819200 Samples
   > ======================================================================
   > 
   > [Mahout] Full Pipeline (DataLoader -> GPU)...
   >   IO + Encode Time: 70.9394 s
   >   Total Time: 70.9394 s (11547.9 vectors/sec)
   > 
   > ======================================================================
   > THROUGHPUT (Higher is Better)
   > Samples: 819200, Qubits: 16
   > ======================================================================
   > Mahout          11547.9 vectors/sec
   > ----------------------------------------------------------------------
   > ```
   > 
   > After:
   > 
   > ```
   > $ python qdp-python/benchmark/benchmark_throughput.py \
   >   --qubits 16 --batches 3200 --batch-size 256 --prefetch 32 \
   >   --frameworks mahout --encoding-method angle
   > Generating 819200 samples of 16 qubits...
   >   Batch size   : 256
   >   Vector length: 65536
   >   Batches      : 3200
   >   Prefetch     : 32
   >   Frameworks   : mahout
   >   Encode method: angle
   >   Generated 819200 samples
   >   PennyLane/Qiskit format: 409600.00 MB
   >   Mahout format: 409600.00 MB
   > 
   > ======================================================================
   > DATALOADER THROUGHPUT BENCHMARK: 16 Qubits, 819200 Samples
   > ======================================================================
   > 
   > [Mahout] Full Pipeline (DataLoader -> GPU)...
   >   IO + Encode Time: 70.8684 s
   >   Total Time: 70.8684 s (11559.4 vectors/sec)
   > 
   > ======================================================================
   > THROUGHPUT (Higher is Better)
   > Samples: 819200, Qubits: 16
   > ======================================================================
   > Mahout          11559.4 vectors/sec
   > ----------------------------------------------------------------------
   > ```
   
   updated the results


-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to