{ "seed": 1, "decoder": { "unets": [ { "dim": 256, "image_embed_dim": 768, "channels": 3, "dim_mults": [1, 2, 3, 4] } ], "image_sizes": [64], "channels": 3, "timesteps": 1000, "loss_type": "l2", "beta_schedule": ["cosine"], "learned_variance": true }, "data": { "webdataset_base_url": "pipe:s3cmd get s3://s-datasets/laion5b/laion2B-data/{}.tar -", "embeddings_url": "s3://s-laion5b/embeddings/vit-l-14/laion2B-en/img_emb_reordered_tiny/", "num_workers": 4, "batch_size": 24, "start_shard": 0, "end_shard": 49, "shard_width": 6, "index_width": 4, "splits": { "train": 0.75, "val": 0.15, "test": 0.1 }, "shuffle_train": false, "resample_train": true, "preprocessing": { "RandomResizedCrop": { "size": [64, 64], "scale": [0.75, 1.0], "ratio": [1.0, 1.0] }, "ToTensor": true } }, "train": { "epochs": 100, "lr":5e-5, "wd": 0.1, "max_grad_norm": 0.5, "save_every_n_samples": 1000, "n_sample_images": 10, "device": "cuda:0", "epoch_samples": 1500, "validation_samples": 1000, "use_ema": true, "ema_beta": 0.99, "unet_training_mask": [true] }, "evaluate": { "n_evaluation_samples": 30, "FID": { "feature": 64 }, "LPIPS": { "net_type": "vgg", "reduction": "mean" } }, "tracker": { "data_path": "/opt/dalle2_generator/test_models", "overwrite_data_path": true, "log": { "log_type": "wandb", "wandb_entity": "veldrovive", "wandb_project": "trackers_tests", "wandb_run_name": "test", "verbose": true }, "load": { "load_from": "wandb", "wandb_run_path": "veldrovive/dalle2_train_decoder/2gs6020x", "wandb_file_path": "latest.pth" }, "save": [{ "save_to": "wandb" }, { "save_to": "huggingface", "huggingface_repo": "Veldrovive/test_model", "save_all": false, "save_latest": true, "save_best": true, "save_type": "model" }] } }