Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Hi, I use the following configuration for training, but when training epoch21, the accuracy has been 60.00 to 60.01 does not rise #257

Open
leo23ui opened this issue Jan 13, 2025 · 0 comments

Comments

@leo23ui
Copy link

leo23ui commented Jan 13, 2025

Hi, I use the following configuration for training, but when training epoch21, the accuracy has been 60.00 to 60.01 does not rise, Is this because I did not use 8*512 batchsize for training? could you please give me some advice, thanks!!!!

export NNODES=1
export GPUS_PER_NODE=2
export WANDB__SERVICE_WAIT=60
export CUDA_VISIBLE_DEVICES=4,5

DISTRIBUTED_ARGS="--nproc_per_node $GPUS_PER_NODE --nnodes $NNODES "
torchrun $DISTRIBUTED_ARGS src/training/main2.py
--save-frequency 1
--report-to wandb
--train-data /home/gg/gg/MQBench-main/test/model/e1/split_2tar
--dataset-type webdataset
--imagenet-val ./ImageNet
--warmup 2000
--batch-size 2048
--epochs 25
--workers 16
--model TinyCLIP-ViT-39M-16-Text-19M
--name exp_name2
--seed 0
--local-loss
--grad-checkpointing
--output ./outputs/c
--lr 0.0001
--gather-with-grad
--pretrained-image-file ViT-B-16@openai
--pretrained-text-file ViT-B-16@openai
--distillation-teacher ViT-B-32@laion2b_e16
--norm_gradient_clip 5
--train-num-samples 15000000
--logit-scale 50

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant