import os import sys sys.path.append(os.path.dirname(os.path.abspath(__file__)) + '/..') import argparse import importlib parser = argparse.ArgumentParser(description='') parser.add_argument('--name', type=str) parser.add_argument('--config', type=str) parser.add_argument('--debug', action='store_true') args = vars(parser.parse_args()) config_module = importlib.import_module(args['config']) config = config_module.config if args['debug']: from configs.ae_curv_cape_m import config os.environ["CUDA_DEVICE_ORDER"]="PCI_BUS_ID" os.environ["CUDA_VISIBLE_DEVICES"] = config['gpu'] if args['debug']: os.environ["CUDA_LAUNCH_BLOCKING"] = '1' import time import wandb import random import numpy as np import torch import cv2 # from torchmetrics import MultiScaleStructuralSimilarityIndexMeasure as MSSSIM from custom_utils.custom_utils import UV2Mesh, get_summary_html, maskFacesBatch, get_summary_str import plotly.graph_objects as go import plotly.express as ex import torch.optim as optim import cmapy as cm if config['is_skirt']: from losses.losses import edge_loss_skirt as edge_loss else: from losses.losses import edge_loss from data.dataset import Dataset # from torch.utils.data import DataLoader from custom_utils.custom_utils import plot_3d from models.cond_encoder import AutoEncoder from utils.checkpoints import CheckpointSaver, load_model from custom_utils.custom_utils import Accumulator from utils.utils import InfiniteDataLoader as DataLoader from torchmetrics.classification import MulticlassJaccardIndex from torchmetrics import Dice from pytorch_msssim import MS_SSIM def init_wandb(): run = wandb.init(project='ae', name=config['name'], notes=config['notes'], dir='/data-local/127/temp', tags=config['tags']) artifact = wandb.Artifact('code', type='code') artifact.add_file('scripts/ae_curv_train.py') artifact.add_file('losses/losses.py') artifact.add_file('models/base.py') artifact.add_file('models/cond_encoder.py') artifact.add_file('models/convnext.py') artifact.add_file('data/dataset.py') config_file_names = [key for key in sys.modules.keys() if key.startswith('configs.')] for config_file_name in config_file_names: file_name = '/'.join(config_file_name.split('.')) + '.py' if os.path.exists(file_name): artifact.add_file(file_name) run.log_artifact(artifact) uv2mesh = UV2Mesh(res=256, is_skirt=config['is_skirt'], is_smpl_orig=config['is_cape']) def train_step(batch): model.train() pred = model([batch['uv_curvature'], batch['uv_mask']]) gt_dense = torch.argmax(batch['uv_curvature'], dim=1) ce_uv_loss = _ce_uv(pred, gt_dense) mesh_pred = uv2mesh(pred) * batch['mesh_mask'][:,:,None] mesh_orig = uv2mesh(gt_dense) * batch['mesh_mask'] ce_mesh_loss = _ce_mesh(torch.transpose(mesh_pred, 1, 2), mesh_orig.long()) loss = ce_uv_loss + ce_mesh_loss optimizer.zero_grad() loss.backward() optimizer.step() model.eval() with torch.inference_mode(): ms_ssim = torch.mean(_MSSSIM(torch.argmax(pred, dim=1)[:,None,:,:].float(), gt_dense[:,None,:,:].float())) mean_iou = _mean_iou(pred, gt_dense) dice = _dice(pred, gt_dense) return { 'train/loss': loss, 'train/ce_uv_loss': ce_uv_loss, 'train/ssim': ms_ssim, 'train/mean_iou': mean_iou, 'train/dice': dice, 'train/ce_mesh_loss': ce_mesh_loss } def valid_step(batch): model.eval() with torch.inference_mode(): pred = model([batch['uv_curvature'], batch['uv_mask']]) gt_dense = torch.argmax(batch['uv_curvature'], dim=1) ce_uv_loss = _ce_uv(pred, gt_dense) mesh_pred = uv2mesh(pred) * batch['mesh_mask'][:,:,None] mesh_orig = uv2mesh(gt_dense) * batch['mesh_mask'] ce_mesh_loss = _ce_mesh(torch.transpose(mesh_pred, 1, 2), mesh_orig.long()) loss = ce_uv_loss + ce_mesh_loss ms_ssim = torch.mean(_MSSSIM(torch.argmax(pred, dim=1)[:,None,:,:].float(), gt_dense[:,None,:,:].float())) mean_iou = _mean_iou(pred, gt_dense) dice = _dice(pred, gt_dense) return { 'valid/loss': loss, 'valid/ce_uv_loss': ce_uv_loss, 'valid/ssim': ms_ssim, 'valid/mean_iou': mean_iou, 'valid/dice': dice, 'valid/ce_mesh_loss': ce_mesh_loss } def plot_everything(batch: dict, type: str): model.eval() pred = model([batch['uv_curvature'], batch['uv_mask']]) # gt_dense = torch.argmax(batch['uv_curvature'], dim=1) # mesh_pred = uv2mesh(pred) * batch['mesh_mask'][:,:,None] # mesh_orig = uv2mesh(gt_dense) * batch['mesh_mask'] orig_img = torch.argmax(batch['uv_curvature'], dim=1) pred_img = torch.argmax(pred, dim=1) img = torch.concat((orig_img, pred_img), axis=2).cpu().numpy() imgs = [wandb.Image((img[i] + 1) / 2.0) for i in range(4)] colors = np.array([cm.color('jet', 1 - i) for i in np.linspace(0, 1, 6+1)]) color_img = np.zeros((img.shape[0], img.shape[1], img.shape[2], 3), np.uint8) for i in range(7): color_img[img == i, :] = colors[i] color_img = [wandb.Image(color_img[i]) for i in range(4)] imgs = { type + '/images': [wandb.Image(img[i,:,:,np.newaxis] / 6.0) for i in range(img.shape[0])], type + '/color_images': color_img } return imgs def train(): valid_iterator = iter(valid_dataloader) step = 0 for batch in train_dataloader: train_loss = train_step(batch) train_accumulator.update(train_loss) if step % 25 == 0: data_valid = next(valid_iterator) valid_loss = valid_step(data_valid) valid_accumulator.update(valid_loss) if step % 1000 == 0 or (step < 1000 and step % 100 == 0): to_log = dict({ **train_loss, **valid_loss, **{ 'current_step': step } }) if step % 5000 == 0 or (step < 5000 and step % 1000 == 0): train_figs = plot_everything(batch, 'train') valid_figs = plot_everything(data_valid, 'valid') to_log = { **to_log, **train_figs, **valid_figs } if step == 0: to_log = { **to_log, **{ 'model/summary': wandb.Html(get_summary_html(model)) } } if not args['debug']: wandb.log(to_log, step=step) train_accumulator.reset() valid_accumulator.reset() print(step) if step % 10000 == 0: ckpt_saver_enc(model.encoder, step, optimizer, None) ckpt_saver_dec(model.decoder, step, optimizer, None) step += 1 if __name__ == '__main__': if not args['debug']: init_wandb() skirt_modifier = '_skirt' if config['is_skirt'] else '' uv_mask_body = np.load('custom_utils/mask_256x512' + ('_skirt' if config['is_skirt'] else '') + '.npy') uv_mask_body = cv2.dilate(uv_mask_body.astype(np.float32), cv2.getStructuringElement(cv2.MORPH_RECT, (2 * 2 + 1, 2 * 2 + 1), (2, 2)), iterations=2) uv_mask_body = torch.from_numpy(uv_mask_body).float().to(config['device']) dataset_train = Dataset(config['train_dir'], config['signature'], device=config['device'], frames_file=config['train_frames'], use_cache=config['dataset_in_memory'], is_cape=config['is_cape']) dataset_valid = Dataset(config['valid_dir'], config['signature'], device=config['device'], frames_file=config['valid_frames'], use_cache=config['dataset_in_memory'], is_cape=config['is_cape']) train_dataloader = DataLoader(dataset_train, batch_size=4, drop_last=True, shuffle=True) valid_dataloader = DataLoader(dataset_valid, batch_size=4, drop_last=True, shuffle=True) train_accumulator = Accumulator() valid_accumulator = Accumulator() model : AutoEncoder = AutoEncoder(config['cond_ae']) optimizer = optim.Adam(model.parameters(), lr=config['lr']) # _MSSSIM = MSSSIM(data_range=2).to(config['device']) _MSSSIM = MS_SSIM(data_range=2, size_average=True, channel=1) _mean_iou = MulticlassJaccardIndex(num_classes=7).to(config['device']) _dice = Dice(average='macro', num_classes=7).to(config['device']) _ce_uv = torch.nn.CrossEntropyLoss() _ce_mesh = torch.nn.CrossEntropyLoss() if config['pretrained_model'] != '': load_model(model.encoder, config['pretrained_model'] + '_enc') load_model(model.decoder, config['pretrained_model'] + '_dec') ckpt_saver_enc = CheckpointSaver('checkpoints/iccv/torch/' + config['name'] + '_enc', top_n=5) ckpt_saver_dec = CheckpointSaver('checkpoints/iccv/torch/' + config['name'] + '_dec', top_n=5) run_name = config['name'] print(f'started training: {run_name}') if not args['debug']: # model = torch.compile(model) wandb.watch(model, log='all', log_freq=1000, log_graph=True) train()