OOMkilled atepoch 99of 100Test dataleaked intotraining setThe datais toonoisy touseRestartkernel,run all,prayWhoannotatedthis?!Worksperfectlyon mymachineLoss wentto NaN onstep 1Servercrashedduring thelive demoCUDAout ofmemory“Quick 5-min sync”= 45minutesThe serverissuspiciouslyslow todayJust addONE morefeaturebeforelaunchForgot toshuffle thedatasetWait…which celldid I runlast?The labelsarecompletelywrongI’lldocumentthis laterCan wemake itreal-time?Serverrebootedovernight.RIP myresults.Stakeholderghostedafter week2Forgot tokill myprocess…sorry team!Datasethas 47samples.Total.Petphotobombedthe video callExtremeclassimbalancestrikes againActually,let’s pivotthe entireapproachFree!SSHconnectiondied mid-experimentInternetdied atdemotimeWe needmoredataThisshould besimple,right?Nicepajamatop on thecallThisnotebookis 2000cells longWho lefttrainingrunning ALLweekend?A meetingcouldhave beenan emailRequirementschanged…againSomeonehoggedall GPUsThebusinessneed has…evolvedJupyterkernel diedmysteriouslyAs permy lastemail…Groundtruth isn’tactuallytrueCan weget 99.9%accuracy?When willthe model beproduction-ready?OOMkilled atepoch 99of 100Test dataleaked intotraining setThe datais toonoisy touseRestartkernel,run all,prayWhoannotatedthis?!Worksperfectlyon mymachineLoss wentto NaN onstep 1Servercrashedduring thelive demoCUDAout ofmemory“Quick 5-min sync”= 45minutesThe serverissuspiciouslyslow todayJust addONE morefeaturebeforelaunchForgot toshuffle thedatasetWait…which celldid I runlast?The labelsarecompletelywrongI’lldocumentthis laterCan wemake itreal-time?Serverrebootedovernight.RIP myresults.Stakeholderghostedafter week2Forgot tokill myprocess…sorry team!Datasethas 47samples.Total.Petphotobombedthe video callExtremeclassimbalancestrikes againActually,let’s pivotthe entireapproachFree!SSHconnectiondied mid-experimentInternetdied atdemotimeWe needmoredataThisshould besimple,right?Nicepajamatop on thecallThisnotebookis 2000cells longWho lefttrainingrunning ALLweekend?A meetingcouldhave beenan emailRequirementschanged…againSomeonehoggedall GPUsThebusinessneed has…evolvedJupyterkernel diedmysteriouslyAs permy lastemail…Groundtruth isn’tactuallytrueCan weget 99.9%accuracy?When willthe model beproduction-ready?

AI lab BINGO - Call List

(Print) Use this randomly generated list as your call list when playing the game. There is no need to say the BINGO column name. Place some kind of mark (like an X, a checkmark, a dot, tally mark, etc) on each cell as you announce it, to keep track. You can also cut out each item, place them in a bag and pull words from the bag.


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
  1. OOM killed at epoch 99 of 100
  2. Test data leaked into training set
  3. The data is too noisy to use
  4. Restart kernel, run all, pray
  5. Who annotated this?!
  6. Works perfectly on my machine
  7. Loss went to NaN on step 1
  8. Server crashed during the live demo
  9. CUDA out of memory
  10. “Quick 5-min sync” = 45 minutes
  11. The server is suspiciously slow today
  12. Just add ONE more feature before launch
  13. Forgot to shuffle the dataset
  14. Wait… which cell did I run last?
  15. The labels are completely wrong
  16. I’ll document this later
  17. Can we make it real-time?
  18. Server rebooted overnight. RIP my results.
  19. Stakeholder ghosted after week 2
  20. Forgot to kill my process… sorry team!
  21. Dataset has 47 samples. Total.
  22. Pet photobombed the video call
  23. Extreme class imbalance strikes again
  24. Actually, let’s pivot the entire approach
  25. Free!
  26. SSH connection died mid-experiment
  27. Internet died at demo time
  28. We need more data
  29. This should be simple, right?
  30. Nice pajama top on the call
  31. This notebook is 2000 cells long
  32. Who left training running ALL weekend?
  33. A meeting could have been an email
  34. Requirements changed… again
  35. Someone hogged all GPUs
  36. The business need has… evolved
  37. Jupyter kernel died mysteriously
  38. As per my last email…
  39. Ground truth isn’t actually true
  40. Can we get 99.9% accuracy?
  41. When will the model be production-ready?