Activity

1. Question406 commented on Mar 4, 2024

   

   Question406

   on Mar 4, 2024

   Collaborator

   More actions

   Hi, this is not an error. The code prints this error to indicate that some generations are not valid. All invalid generation is not included in the result folder.
2. kd-scki3011 commented on Mar 5, 2024

   

   kd-scki3011

   on Mar 5, 2024

   Author

   More actions

   Hi, this is not an error. The code prints this error to indicate that some generations are not valid. All invalid generation is not included in the result folder.

   thanks!
3. kd-scki3011 commented on Mar 5, 2024

   

   kd-scki3011

   on Mar 5, 2024

   Author

   More actions

   Hi, this is not an error. The code prints this error to indicate that some generations are not valid. All invalid generation is not included in the result folder.

   
   Hello, excuse me. I would like to ask if the pictures produced during my training are all like this, and the progress is still at about 29%, may I ask where I set it wrong?
   
   this is where I put runwayml/stable-diffusion-inpainting in
4. kd-scki3011 commented on Mar 5, 2024

   

   kd-scki3011

   on Mar 5, 2024

   Author

   More actions

   Hi, this is not an error. The code prints this error to indicate that some generations are not valid. All invalid generation is not included in the result folder.

   Hello, excuse me. I would like to ask if the pictures produced during my training are all like this, and the progress is still at about 29%, may I ask where I set it wrong? this is where I put runwayml/stable-diffusion-inpainting in

   this is my config:

   train on combination of all datas

   data:
   target: "src.trainers.WrappedDataModule"
   batch_size: 1
   scene_data: ocr-dataset/
   synth_data: ocr-dataset/synth/
   train:
   size: 256
   max_num: 2000000 # diffly choose this number
   augconf:
   synth:
   center: 0.1
   pad: false
   scene:
   expand_mask:
   center_mask: 0.6
   additional_mask: 0.4
   crop:
   mask_image_ratio: 15
   rotate:
   cat_prob: [1, 0, 0]
   angle_list: [-15, -30, -45, -60, -90, 15, 30, 45, 60, 90]
   rotate_range: 90

   dataconfs:
     ArT:
       type: scene
       label_path: ArT/train_labels.json
       image_dir: ArT/train_images/
   
     #COCO:
     #  type: scene
     #  label_path: ocr-dataset/COCO/cocotext.v2.json
     #  image_dir: ocr-dataset/COCO/train2014/
   
     #TextOCR:
     #  type: scene
     #  label_path: TextOCR/TextOCR_0.1_train.json
     #  image_dir: TextOCR/train_images/
   
     #Synthtiger:
     #  type: synth
     #  label_path: ocr-dataset/synth/data.csv
     #  image_dir: ocr-dataset/synth/
     #  style_mode: same-same
     #  use_textbbox: false
     #  style_dropout: [0.5, 0.5]
     #  rand_mask_text: true
   

   validation:
   size: 256
   # max_num: 6400 # diffly choose this number
   augconf:
   synth:
   center: 1.
   pad: false
   scene:
   expand_mask:
   center_mask: 0.
   additional_mask: 0.
   crop:
   mask_image_ratio: 30
   rotate:
   cat_prob: [1, 0, 0]
   angle_list: [-15, -30, -45, -60, -90, 15, 30, 45, 60, 90]
   rotate_range: 90

   dataconfs:
     #ArT:
     #  type: scene
     #  label_path: ArT/train_labels.json
     #  image_dir: ArT/train_images/
   
     #COCO:
     #  type: scene
     #  label_path: ocr-dataset/COCO/cocotext.v2.json
     #  image_dir: ocr-dataset/COCO/train2014/
   
     TextOCR:
       type: scene
       label_path: TextOCR/TextOCR_0.1_val.json
       image_dir: TextOCR/train_images/
   

   model:
   source: raw
   target: "src.trainers.CharInpaintModelWrapper"
   pretrained_model_path: runwayml/stable-diffusion-inpainting
   loss_type: MaskMSELoss
   loss_alpha: 5
   base_learning_rate: 5.0e-5
   precision: 16
   weight_decay: 0.0
   adam_epsilon: 1.0e-8
   freeze_char_embedder: false
   optimize_vae: false
   vae:

   tokenizer:
   model_max_length: 20
   char_tokenizer:
   pretrained_path: checkpoints/chartokenizer
   pad_token: " "
   unk_token: " "
   model_max_length: 20
   char_embedder:
   vocab_size: 95 # by default
   embedding_dim: 32
   max_length: 20
   padding_idx: 0
   attention_head_dim: 2
   unet:
   attention_head_dim: { "text": 8, "char": 2 }
   cross_attention_dim: { "text": 768, "char": 32 }
   noise_scheduler: diffusers.DDIMScheduler

   lightning:
   logger:
   callbacks:
   checkpoint_callback:
   params:
   save_top_k: -1
   image_logger:
   target: "src.trainers.CharInpaintImageLogger"
   params:
   # train_batch_frequency: 2400
   # valid_batch_frequency: 500
   train_batch_frequency: 2
   valid_batch_frequency: 2
   disable_wandb: true
   generation_kwargs:
   num_inference_steps: 30
   num_sample_per_image: 3
   guidance_scale: 7.5
   seed: 42

   # NOTE: Download pretrained ABINet model from https://github.com/FangShancheng/ABINet.git and
   #       put model checkpoints in checkpoints/abinet to use this callback
   # ocracc_logger:
   #   target: "src.trainers.OCRAccLogger"
   #   params:
   #     train_eval_conf:
   #       size: 256
   #       augconf: ${data.validation.augconf}
   #       max_num: 5
   #       dataconfs:
   #         TextOCR:
   #           type: scene
   #           label_path: ${data.scene_data}/TextOCR/TextOCR_0.1_train.json
   #           image_dir: ${data.scene_data}/TextOCR/train_images/
   #           len_counter:
   #             eachnum: 10
   
   #     val_eval_conf:
   #       size: 256
   #       augconf: ${data.validation.augconf}
   #       max_num: 5
   #       dataconfs:
   #         TextOCR:
   #           type: scene
   #           label_path: ${data.scene_data}/TextOCR/TextOCR_0.1_val.json
   #           image_dir: ${data.scene_data}/TextOCR/train_images/
   #           max_num: 1000
   #     base_log_dir: ${base_log_dir}/ocrlogs # will be set in code
   

   trainer:
   accelerator: gpu
   devices: [0]
   strategy: ddp
   amp_backend: native
   log_every_n_steps: 16 # this is global step
   precision: 16
   max_epochs: 15
   check_val_every_n_epoch: 1
   accumulate_grad_batches: 8
   gradient_clip_val: 3.
   gradient_clip_algorithm: norm
   benchmark: true