app.py

import os
import gradio as gr

import wandb
from huggingface_hub import HfApi

TOKEN = os.environ.get("DATACOMP_TOKEN")
API = HfApi(token=TOKEN)
wandb_api_key = os.environ.get('wandb_api_key')
wandb.login(key=wandb_api_key)

random_num = f"10.0"
subset = f"frac-1over4"
experiment_name = f"ImageNetTraining10.0-frac-1over4"
experiment_repo = f"datacomp/ImageNetTraining10.0-frac-1over4"

def start_train(status_box):
    os.system("echo '#### pwd'")
    os.system("pwd")
    os.system("echo '#### ls'")
    os.system("ls")
    # Create a place to put the output.
    os.system("echo 'Creating results output repository in case it does not exist yet...'")
    try:
        API.create_repo(repo_id=f"datacomp/ImageNetTraining10.0-frac-1over4", repo_type="dataset",)
        os.system(f"echo 'Created results output repository datacomp/ImageNetTraining10.0-frac-1over4'")
    except:
        os.system("echo 'Already there; skipping.'")
        pass
    os.system("echo 'Beginning processing.'")
    # Handles CUDA OOM errors.
    os.system(f"export PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True")
    os.system("echo 'Okay, trying training.'")
    API.add_space_variable(repo_id=experiment_repo, key="STATUS", value="TRAINING")
    os.system(f"cd pytorch-image-models; ./train.sh 4 --dataset hfds/datacomp/imagenet-1k-random-10.0-frac-1over4 --log-wandb --wandb-project ImageNetTraining10.0-frac-1over4 --experiment ImageNetTraining10.0-frac-1over4 --model seresnet34 --sched cosine --epochs 150 --warmup-epochs 5 --lr 0.4 --reprob 0.5 --remode pixel --batch-size 256 --amp -j 4")
    API.add_space_variable(repo_id=experiment_repo, key="STATUS", value="IDLE")
    os.system("echo 'Done'.")
    os.system("ls")
    # Upload output to repository
    os.system("echo 'trying to upload...'")
    API.upload_folder(folder_path="/app", repo_id=f"datacomp/ImageNetTraining10.0-frac-1over4", repo_type="dataset",)
    API.pause_space(experiment_repo)

def run():
    with gr.Blocks() as app:
        gr.Markdown(f"Randomization: 10.0")
        gr.Markdown(f"Subset: frac-1over4")
        status_box = gr.Textbox()
        #space_variables = API.get_space_variables(repo_id=experiment_repo)
        #if 'STATUS' not in space_variables or space_variables['STATUS'].value != 'TRAINING':
        #    API.add_space_variable(repo_id=experiment_repo, key="STATUS", value="IDLE")
        #    status_box.value = "Status: Idle"
        start = gr.Button("Start")
        start.click(start_train, inputs=status_box, outputs=status_box)
        #else:
        #    status_box.value = "Status: Training"
    app.launch(server_name="0.0.0.0", server_port=7860)
    
if __name__ == '__main__':
    run()