A place for holding images

[leondgarse]

TOC

• Model surgery
• COCO
• ImageNet training
• Model architecture
• Visualizing
• CLIP and Text
• Stable Diffusion
• Segmentation models
  

[leondgarse]

Model surgery

• split_conv2d
  

[leondgarse]

COCO

• effdetd0_dog_cat
  
• effdetd1_dynamic_dog_cat
  
• yoloxs_dog_cat
  
• yoloxtiny_dynamic_dog_cat
  
• yolor_csp_dog_cat
  
• yolor_csp_dynamic_dog_cat
  
• yolov7_csp_dog_cat
  
• yolov7_csp_dynamic_dog_cat
  
• yolov8_n_dog_cat
  
• yolov8_s_dynamic_dog_cat
  
• coco_data_aug
  
• random_crop_mode
  

[leondgarse]

ImageNet training

• aotnet50_imagenet
  
• efficientnetv2b0 progressive_cifar10
  
• aotnet50_progressive_160
  
• aotnet_adamw
  
• coatnet0_160
  
• coatnet0_ft_224
  
• coatnet0_224_300
  
• cmt_tiny
  
• cmt_tiny_ft_224
  
• volo_cifar10
  
• swinv2_tiny_ns
  

COCO training

[leondgarse]

Model architecture

• model_summary
  
• rel_to_abs
  
• convnext
  
• efficientdet
  
• cmt
  
• coatnet
  
• halo_attention
  
• uniformer
  
• nat
  
• efficientformer
  
• hornet
  
• ghostnetv2
  
• fasternet
  

[leondgarse]

Visualizing

• aotnet50_imagenet_parallel_coordinates
  
• resnest50_visualize_filters
  
• resnest50_make_gradcam_heatmap
  
• resnest50_make_and_apply_gradcam_heatmap
  
• resnest50_make_and_apply_gradcam_heatmap_dog
  
• resnest50_make_and_apply_gradcam_heatmap_cat
  
• beit_plot_attention_score_maps
  
• levit_plot_attention_score_maps
  
• botnet_plot_attention_score_maps
  
• halonet_plot_attention_score_maps
  
• coatnet_plot_attention_score_maps
  
• coat_plot_attention_score_maps
  
• vit_plot_attention_score_maps
  

[leondgarse]

CLIP and Text

• clip_torch_tf
  
• clip_out
  
• clip_out_3
  
• text_tf_torch
  

[leondgarse]

Stable Diffusion

• stable_diffusion
  
• stable_diffusion_512_1024
  
• stable_diffusion_384_768
  
• stable_diffusion_inner
  
• stable_duffusion_image_2_iamge
  
• stable_diffusion_inpaint
  
• ddpm_unet_test_torch_E200 train
  
• ddpm_unet_test_E100
  
• gtsrb_15
  
• ddpm_unet_test_torch_E200 predict
  

[leondgarse]

Segmentation models

• yolov8_s_dynamic_segment_dog_cat
  
• sam_mobile_sam_5m_raw
  
• sam_mobile_sam_5m
  
• sam_efficientvit_l0_box
  
• sam_efficientvit_l0_yolov8n_box
  
• sam_efficientvit_l0 torch