Commits

Branch selector

 main

User selector

All users

Datepicker

All time

Commit History

Commits on Dec 7, 2022

• Update README.md

  justinpinkney

  authored
  ef843a0

  Copy full SHA for ef843a0

Commits on Nov 24, 2022

• Create LICENSE

  justinpinkney

  authored
  e9cc3cb

  Copy full SHA for e9cc3cb

Commits on Nov 1, 2022

• Merge pull request #12 from osanseviero/patch-1

  Show description for 56b71c4
  justinpinkney

  authored
  56b71c4

  Copy full SHA for 56b71c4

Commits on Oct 27, 2022

• Remove unneeded set_format

  osanseviero

  authored
  81ed22d

  Copy full SHA for 81ed22d

Commits on Oct 17, 2022

• update spaces link

  justinpinkney

  authored
  4d4386e

  Copy full SHA for 4d4386e

Commits on Oct 11, 2022

• Merge pull request #10 from LambdaLabsML/eole/benchmarking-deployment

  Show description for 68018b0
  eolecvk

  authored
  68018b0

  Copy full SHA for 68018b0

• title levels fix

  eolecvk

  committed
  d4e85c7

  Copy full SHA for d4e85c7

• add dockerfile at root and update docs/benchmark to include docker instructions

  eolecvk

  committed
  1233af7

  Copy full SHA for 1233af7

• Merge pull request #7 from LambdaLabsML/eole/fp16-v-autocast

  Show description for 4c74b72
  eolecvk

  authored
  4c74b72

  Copy full SHA for 4c74b72

• add vscode to gitignore

  eolecvk

  committed
  cc801b8

  Copy full SHA for cc801b8

• remove erroneous import

  eolecvk

  committed
  5cf05f4

  Copy full SHA for 5cf05f4

Commits on Oct 7, 2022

• oops README benchmark-deploy not in right branch

  eolecvk

  authored
  dfbacce

  Copy full SHA for dfbacce

• oops Dockerfile committed but not the right branch

  eolecvk

  authored
  cbd7014

  Copy full SHA for cbd7014

• cuda autocast y/n argument

  eolecvk

  committed
  2c1e19e

  Copy full SHA for 2c1e19e

• rm benchmark.tar

  eolecvk

  committed
  90bd1b4

  Copy full SHA for 90bd1b4

• csv filepath typo

  eolecvk

  committed
  e5979ac

  Copy full SHA for e5979ac

Commits on Oct 6, 2022

• run container in detach mode for containerid variable assignment to work

  eolecvk

  committed
  7d98573

  Copy full SHA for 7d98573

• benchmark docker somewhat streamlined, w README

  eolecvk

  committed
  ad31ad6

  Copy full SHA for ad31ad6

• benchmarking deployment

  eolecvk

  committed
  3bc6413

  Copy full SHA for 3bc6413

Commits on Oct 5, 2022

• Merge pull request #3 from LambdaLabsML/eole/benchmarking

  Show description for fe172cb
  chuanli11

  authored
  fe172cb

  Copy full SHA for fe172cb

• missing csv header 'runtime'

  eolecvk

  committed
  2b5da79

  Copy full SHA for 2b5da79

• saving logs to benchmark_tmp.csv instead of benchmark.csv and gitignore _tmp.csv; better try/catch if missing ACCESS TOKEN

  eolecvk

  committed
  645fc33

  Copy full SHA for 645fc33

• benchmark: change latency to speed/secs to finish in figure title

  chuanli11

  committed
  5245735

  Copy full SHA for 5245735

Commits on Oct 4, 2022

• rm deprecated graphs .svg

  eolecvk

  committed
  d56c0c5

  Copy full SHA for d56c0c5

• update latency graph rtx8000, update readme.md results section

  eolecvk

  committed
  0c9fbc7

  Copy full SHA for 0c9fbc7

• typos

  eolecvk

  committed
  437689b

  Copy full SHA for 437689b

• update benchmark.md prose and prettify bar graphs

  eolecvk

  committed
  7f6ced3

  Copy full SHA for 7f6ced3

Commits on Oct 3, 2022

• benchmark: Add readme to benchmark

  chuanli11

  committed
  cc81843

  Copy full SHA for cc81843

• benchmark: add revsiion when loading pretrianed models for half precision benchmark

  chuanli11

  committed
  2469d61

  Copy full SHA for 2469d61

Commits on Oct 2, 2022

• benchmark: use autocast in a more elegant way. remove onnx from gpu benchmark

  chuanli11

  committed
  37b5afd

  Copy full SHA for 37b5afd

• benchmark: more results

  chuanli11

  committed
  c6598a5

  Copy full SHA for c6598a5

• benchmark: handles onnx runtime error

  chuanli11

  committed
  254536f

  Copy full SHA for 254536f

Commits on Oct 1, 2022

• benchmark: use float instead of str for logging

  chuanli11

  committed
  e873162

  Copy full SHA for e873162

• benchmark: 1) Catch CUDA OOM error more gracefully 2) Add samples (batch size) as a cmd line argument.

  chuanli11

  committed
  91542b2

  Copy full SHA for 91542b2

• benchmark: update requirements.txt so it uses torch cu116 build and onnxruntime-gpu

  chuanli11

  committed
  03df267

  Copy full SHA for 03df267

Pagination

Previous

Next