docs / benchmarks / ml

Benchmarks index
  1. Overview
  1. Exemptions
  2. Methodology

aot

  1. Overview
  2. Components

codec

  1. Overview
  2. Base32
  3. Base64
  4. Deflate (dev.cajeta.codec)

codec / csv

  1. CSV

codec / json

  1. JSON

collection

  1. Overview
  2. ArrayList
  3. BPlusTree
  4. Cache
  5. Collectors
  6. HashMap
  7. HashSet
  8. Heap
  9. ImmutableList
  10. ImmutableMap
  11. ImmutableSet
  12. LinkedList
  13. RedBlackTree
  14. Sort

collection / ltm

  1. LtmBPlusTree

concurrent

  1. Overview
  2. AtomicInt32
  3. AtomicInt64
  4. Channel
  5. Fiber
  6. FiberLocal
  7. Lock
  8. Mutex
  9. RwLock
  10. Semaphore
  11. Tasks

gfx

  1. Overview
  2. Texture2D

hash

  1. Overview
  2. DefaultHasher
  3. Hash
  4. SipHash
  5. XXHash3

http

  1. Overview
  2. Client
  3. Server
  4. Server-sent events
  5. Start-up and footprint
  6. TLS
  7. WebSocket

io

  1. Overview
  2. Buffer

io / file

  1. Directory walk
  2. File
  3. FileInfo
  4. FileReader
  5. FileWriter
  6. Path
  7. Watcher

io / net

  1. BufferPool
  2. ByteBuffer
  3. Server
  4. TcpListener
  5. TcpStream
  6. UdpSocket

io / net / uri

  1. Uri
  2. UriBuilder

lang

  1. Overview
  2. Allocation and ownership
  3. Calls and dispatch
  4. Exceptions
  5. Guid
  6. Loops
  7. Math and SIMD
  8. Optional
  9. Pair
  10. Slice
  11. String
  12. StringBuilder
  13. Whole programs: the Benchmarks Game

lang / stream

  1. ArrayStream
  2. Stream

llm

  1. Overview
  2. Decode
  3. Kernel parity
  4. Load and first token
  5. Long context
  6. Memory
  7. Prefill

math

  1. Overview
  2. Ray
  3. Rotation
  4. Tensor
  5. Transform

math / fft

  1. Fft

math / linalg

  1. LinAlg

math / npio

  1. Npy

math / poly

  1. Poly

math / random

  1. Generator

math / stats

  1. Stats

ml

  1. Overview
  2. Classical models
  3. Neural models
  4. XGBoost

nucleo

  1. Overview
  2. Column
  3. Fuse
  4. Table
  5. Tape
  6. Transforms

primavera

  1. Overview
  2. Build time
  3. Dependency injection at scale
  4. REST endpoint
  5. Start-up and footprint
  6. Templated page

process

  1. Overview
  2. Command
  3. Process

reflect

  1. Overview
  2. Class

search

  1. Overview

search / distance

  1. Distance

search / fuzzy

  1. Matcher

search / ngram

  1. Index

time

  1. Overview
  2. Clock
  3. DateTimeFormatter
  4. Duration
  5. Instant
  6. LocalDate
  7. LocalDateTime
  8. LocalTime
  9. ZonedDateTime

toolchain

  1. Overview
  2. Binary size
  3. Build tool
  4. Compile time
  5. Debugger
  6. Notebook kernel start-up
  7. Start-up

toolchain / kernel-startup

  1. Overview
  2. Where cold-start time goes

wire

  1. Overview
  2. CompressStream
  3. Encoder
  4. SchemaEncoder

xpu

  1. Overview
  2. Device
  3. KernelBuffer
  4. KernelStream
  5. KernelTimer

xpu / mesh

  1. MeshSimplifier
Machine
no runs yet

Neural models

dev.cajeta.ml — the torch role over Tensor<float32>: a training step and inference for an MLP, a small CNN and a transformer block at batch 64, on the CPU and on the device, plus the autodiff overhead and checkpoint load.

Benchmarks

idworkloadagainstcheck
ml/mlp-train-step784-512-10 MLP, one step; variants cpu, devicePyTorch (CPU, CUDA/ROCm)loss decreases over 100 steps
ml/mlp-inferinference at batch 64PyTorchlogits within epsilon
ml/cnn-train-steptwo-conv CNN on 28×28, one stepPyTorchloss decreases
ml/cnn-inferinference at batch 64PyTorchwithin epsilon
ml/transformer-block-trainone encoder block, d = 512, seq 128, one stepPyTorchloss decreases
ml/transformer-block-inferinferencePyTorchwithin epsilon
ml/autodiff-overheadbackward over forward ratio for the MLPPyTorch autogradgradients within epsilon
ml/checkpoint-loadload a torch checkpoint of the MLP, 1e2 timestorch.loadweights equal

Results

One chart per benchmark, for the machine selected at the top of the page. Languages are the bars, best first; Cajeta’s bar is in color and the competitors’ are gray. A skipped competitor keeps its row with the reason in place of a bar. A benchmark with no row on the selected machine says so and shows nothing else.

ml/mlp-train-step

Not yet measured on Proton.
ml/mlp-infer

Not yet measured on Proton.
ml/cnn-train-step

Not yet measured on Proton.
ml/cnn-infer

Not yet measured on Proton.
ml/transformer-block-train

Not yet measured on Proton.
ml/transformer-block-infer

Not yet measured on Proton.
ml/autodiff-overhead

Not yet measured on Proton.
ml/checkpoint-load

Not yet measured on Proton.

Reproduce

cd bench
./bench.sh --bench mlp-train-step               # Cajeta
./bench.sh --bench mlp-train-step --competitors # every pinned competitor

Every row records the compiler and competitor versions, flags, machine, OS and driver; METHODOLOGY states the rules a number must satisfy before it is published.

See also

  • Benchmark source: cajeta-ml/bench/ and cajeta-xgboost/bench/
  • Library: cajeta-ml

Source: bench/ml/Neural.md · 1 min read · 282 words