update

.gitattributes

@@ -0,0 +1,44 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+assets/01.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/02.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/03.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/04.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/05.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/06.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/07.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/08.mp4 filter=lfs diff=lfs merge=lfs -text
+assets/09.mp4 filter=lfs diff=lfs merge=lfs -text

.gitignore

@@ -0,0 +1,4 @@
+data_ssc/
+demo_out/
+pretrained_models/*
+.vscode/

README.md

@@ -0,0 +1,13 @@
+---
+title: AiOS
+emoji: ⚡
+colorFrom: blue
+colorTo: indigo
+sdk: gradio
+python_version: 3.9
+sdk_version: 4.38.1
+app_file: app.py
+pinned: false
+---
+
+Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference

app.py

@@ -0,0 +1,126 @@
+import os
+import sys
+import subprocess
+import pkg_resources
+
+def is_package_installed(package_name):
+    try:
+        pkg_resources.get_distribution(package_name)
+        return True
+    except pkg_resources.DistributionNotFound:
+        return False
+
+if is_package_installed("mmcv"):
+    print("MMCV is installed.")
+else:
+    print("MMCV is not installed. Build it from the source.")
+    os.environ["MMCV_WITH_OPS"] = "1"
+    os.environ["FORCE_MLU"] = "1"
+    subprocess.run(["pip", "install", "-e", "./mmcv"], check=True)
+    subprocess.run(["pip", "list"], check=True)
+
+if is_package_installed("pytorch3d"):
+    print("pytorch3d is installed.")
+else:
+    print("pytorch3d is not installed. Build it from the source.")
+    subprocess.run(["pip", "install", "-e", "./pytorch3d"], check=True)
+
+if is_package_installed("MultiScaleDeformableAttention"):
+    print("MultiScaleDeformableAttention is installed.")
+else:
+    print("MultiScaleDeformableAttention is not installed. Build it from the source.")
+    subprocess.run(["pip", "install", "-e", "./models/aios/ops"], check=True)
+    
+import os.path as osp
+from pathlib import Path
+import cv2
+import gradio as gr
+import torch
+import math
+import spaces
+from huggingface_hub import hf_hub_download
+
+hf_hub_download(repo_id="ttxskk/AiOS", filename="aios_checkpoint.pth", local_dir="/home/user/app/pretrained_models")
+
+OUT_FOLDER = '/home/user/app/demo_out'
+os.makedirs(OUT_FOLDER, exist_ok=True)
+
+DEMO_CONFIG = '/home/user/app/config/aios_smplx_demo.py'
+MODEL_PATH = '/home/user/app/pretrained_models/aios_checkpoint.pth'
+@spaces.GPU(enable_queue=True, duration=300)
+def infer(video_input, batch_size, threshold=0.5, num_person=1):
+    os.system(f'rm -rf {OUT_FOLDER}/*')
+    os.system(f'torchrun --nproc_per_node 1 \
+             main.py \
+            -c {DEMO_CONFIG} \
+            --options batch_size={batch_size} backbone="resnet50" num_person={num_person} threshold={threshold} \
+            --resume {MODEL_PATH} \
+            --eval \
+            --inference \
+            --inference_input {video_input} \
+            --to_vid \
+            --output_dir {OUT_FOLDER}')
+    
+    video_path = os.path.join(OUT_FOLDER, 'demo_vid.mp4')
+    save_path_img = os.path.join(OUT_FOLDER, 'res_img')
+    save_path_mesh = os.path.join(OUT_FOLDER, 'mesh')
+    save_mesh_file = os.path.join(OUT_FOLDER, 'mesh.zip')
+    os.system(f'zip -r {save_mesh_file} {save_path_mesh}')
+    yield video_path, save_mesh_file
+
+TITLE = """
+
+<div style="display: flex; justify-content: center; align-items: center; text-align: center;">
+  <div>
+    <h1 align="center">AiOS: All-in-One-Stage Expressive Human Pose and Shape Estimation</h1>
+  </div>
+</div>
+
+<div style="display: flex; justify-content: center; align-items: center; text-align: center;">
+    <div style="display:flex; gap: 0.25rem;" align="center">
+        <a href="https://ttxskk.github.io/AiOS/" target="_blank"><img src='https://img.shields.io/badge/Project-Page-Green'></a>
+        <a href="https://github.com/ttxskk/AiOS" target="_blank"><img src='https://img.shields.io/badge/Github-Code-blue'></a>
+        <a href="https://ttxskk.github.io/AiOS/assets/aios_cvpr24.pdf" target="_blank"><img src='https://img.shields.io/badge/Paper-Arxiv-red'></a>
+    </div>
+</div>
+<div style="font-size: 1.1rem; color: #555; max-width: 800px; margin: 1rem auto; line-height: 1.5; justify-content: center; align-items: center; text-align: center;">
+  <div>
+    <p>Recover multiple expressive human pose and shape recovery from an RGB image without any additional requirements, such as an off-the-shelf detection model.</h1>
+  </div>
+</div>
+"""
+with gr.Blocks(title="AiOS", theme=gr.themes.Soft(primary_hue="blue", secondary_hue="gray")) as demo:
+    
+    gr.Markdown(TITLE)
+    with gr.Row():
+        with gr.Column(scale=2):
+            video_input = gr.Video(label="Input video", elem_classes="video")
+        with gr.Column(scale=1):
+            batch_size = gr.Textbox(label="Batch Size", type="text", value=8)
+            num_person = gr.Textbox(label="Number of Person", type="text", value=1)
+            threshold = gr.Slider(0, 1.0, value=0.5, label='Score Threshold')
+            send_button = gr.Button("Infer")
+    gr.HTML("""<br/>""")
+        
+    with gr.Row():
+        with gr.Column():
+            # processed_frames = gr.Image(label="Last processed frame")
+            video_output = gr.Video(elem_classes="video")
+        with gr.Column():
+            meshes_output = gr.File(label="3D meshes")
+    
+    send_button.click(fn=infer, inputs=[video_input, batch_size, threshold, num_person], outputs=[video_output, meshes_output])
+    # example_videos = gr.Examples([
+    #     ['./assets/01.mp4'], 
+    #     ['./assets/02.mp4'], 
+    #     ['./assets/03.mp4'],
+    #     ['./assets/04.mp4'], 
+    #     ['./assets/05.mp4'], 
+    #     ['./assets/06.mp4'], 
+    #     ['./assets/07.mp4'], 
+    #     ['./assets/08.mp4'], 
+    #     ['./assets/09.mp4'], 
+    #     ], 
+    #     inputs=[video_input, 0.5])
+    
+demo.queue().launch(debug=True)

assets/01.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ba560996c248d78be6556f1727ae6ced81cd62a002715c3ffd542f6202b204b
+size 2751935

assets/02.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00702a08c978b27b3ddf6ddfd48c5a057753664c8e80d83f4b4e04dff45b8a71
+size 2827267

assets/03.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bfcc1ce90a0921ffa5550a04f743470081ff4599c265cf491e636a8ea70233d4
+size 4033767

assets/04.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:28531c3c0ad9cbcc097a00f8553aafcdc0513a881f0fa6d1a7937248f46fce0c
+size 2639842

assets/05.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1cf7f1b65d87f0a77c1d9456771e4f88228aa836426b4ad0cbad672e80d07e36
+size 3584040

assets/06.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fcb4139d4863c5ec92224f7cb452ec4631be0613eb4c3f82ee7fbb6f89510fe2
+size 19797950

assets/07.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4c71c5ed8573cb727c515d733e51c5da4654c58ab096cbca4bdf9b072e8284c7
+size 3274979

assets/08.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d14f03e984a0ebefd9e8429c8e0d3ecdb0ffc9126ad91a489b57dc0f5d12695b
+size 6825913

assets/09.mp4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:30b5b6f75f024647a9e430f02b33caa1ccec327b487ba5bb451e2859e1e45142
+size 6336699

config/aios_smplx.py

@@ -0,0 +1,259 @@
+
+num_classes = 2
+lr = 0.0001*1.414/10
+param_dict_type = 'default'
+lr_backbone = 1e-05*1.414/10
+lr_backbone_names = ['backbone.0']
+lr_linear_proj_names = ['reference_points', 'sampling_offsets']
+lr_linear_proj_mult = 0.1
+ddetr_lr_param = False
+batch_size = 2
+weight_decay = 0.0001
+epochs = 200
+lr_drop = 11
+save_checkpoint_interval = 1
+clip_max_norm = 0.1
+onecyclelr = False
+multi_step_lr = True
+lr_drop_list = [30, 60]
+
+modelname = 'aios_smplx'
+frozen_weights = None
+backbone = 'resnet50'
+use_checkpoint = False
+
+dilation = False
+position_embedding = 'sine'
+pe_temperatureH = 20
+pe_temperatureW = 20
+return_interm_indices = [1, 2, 3]
+backbone_freeze_keywords = None
+enc_layers = 6
+dec_layers = 6
+pre_norm = False
+dim_feedforward = 2048
+hidden_dim = 256
+dropout = 0.0
+nheads = 8
+num_queries = 900
+query_dim = 4
+num_patterns = 0
+random_refpoints_xy = False
+fix_refpoints_hw = -1
+dec_layer_number = None
+num_feature_levels = 4
+enc_n_points = 4
+dec_n_points = 4
+dln_xy_noise = 0.2
+dln_hw_noise = 0.2
+two_stage_type = 'standard'
+two_stage_bbox_embed_share = False
+two_stage_class_embed_share = False
+two_stage_learn_wh = False
+two_stage_default_hw = 0.05
+two_stage_keep_all_tokens = False
+rm_detach = None
+num_select = 50
+transformer_activation = 'relu'
+batch_norm_type = 'FrozenBatchNorm2d'
+
+masks = False
+losses = ["smpl_pose", "smpl_beta", "smpl_expr",
+        "smpl_kp2d","smpl_kp3d","smpl_kp3d_ra",'labels', 'boxes', "keypoints"]
+# losses = ['labels', 'boxes', "keypoints"]
+aux_loss = True
+set_cost_class = 2.0
+set_cost_bbox = 5.0
+set_cost_giou = 2.0
+set_cost_keypoints = 10.0
+set_cost_kpvis = 0.0
+set_cost_oks = 4.0
+cls_loss_coef = 2.0
+# keypoints_loss_coef = 10.0
+
+smpl_pose_loss_root_coef = 10 * 0.1
+smpl_pose_loss_body_coef = 1 * 0.1
+smpl_pose_loss_lhand_coef = 1 * 0.1
+smpl_pose_loss_rhand_coef = 1 * 0.1
+smpl_pose_loss_jaw_coef = 1 * 0.1
+smpl_beta_loss_coef = 0.01
+smpl_expr_loss_coef = 0.01
+
+# smpl_kp3d_loss_coef = 10
+smpl_body_kp3d_loss_coef = 10.0 * 0.1
+smpl_face_kp3d_loss_coef = 1.0 * 0.1
+smpl_lhand_kp3d_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_loss_coef = 1 * 0.1
+
+# kp3d ra
+smpl_body_kp3d_ra_loss_coef = 10 * 0.1
+smpl_face_kp3d_ra_loss_coef = 1 * 0.1
+smpl_lhand_kp3d_ra_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_ra_loss_coef = 1 * 0.1
+
+
+# smpl_kp2d_ba_loss_coef = 1.0
+smpl_body_kp2d_loss_coef = 10.0 * 0.1
+smpl_lhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_rhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_face_kp2d_loss_coef = 1.0 * 0.1
+
+smpl_body_kp2d_ba_loss_coef = 0 * 0.1
+smpl_face_kp2d_ba_loss_coef = 0 * 0.1
+smpl_lhand_kp2d_ba_loss_coef = 0 * 0.1
+smpl_rhand_kp2d_ba_loss_coef = 0 * 0.1
+
+bbox_loss_coef = 5.0
+body_bbox_loss_coef = 5.0
+lhand_bbox_loss_coef = 5.0
+rhand_bbox_loss_coef = 5.0
+face_bbox_loss_coef = 5.0
+
+giou_loss_coef = 2.0
+body_giou_loss_coef = 2.0
+rhand_giou_loss_coef = 2.0
+lhand_giou_loss_coef = 2.0
+face_giou_loss_coef = 2.0
+
+keypoints_loss_coef = 10.0
+rhand_keypoints_loss_coef = 10.0
+lhand_keypoints_loss_coef = 10.0
+face_keypoints_loss_coef = 10.0
+       
+oks_loss_coef=4.0
+rhand_oks_loss_coef = 0.5
+lhand_oks_loss_coef = 0.5
+face_oks_loss_coef = 4.0
+
+
+enc_loss_coef = 1.0
+interm_loss_coef = 1.0
+no_interm_box_loss = False
+focal_alpha = 0.25
+rm_self_attn_layers = None
+indices_idx_list = [1, 2, 3, 4, 5, 6, 7]
+
+decoder_sa_type = 'sa'
+matcher_type = 'HungarianMatcher'
+decoder_module_seq = ['sa', 'ca', 'ffn']
+nms_iou_threshold = -1
+
+dec_pred_bbox_embed_share = False
+dec_pred_class_embed_share = False
+dec_pred_pose_embed_share = False
+body_only = True
+
+# for dn
+use_dn = True
+dn_number = 100
+dn_box_noise_scale = 0.4
+dn_label_noise_ratio = 0.5
+embed_init_tgt = False
+dn_label_coef = 0.3
+dn_bbox_coef = 0.5
+dn_batch_gt_fuse = False
+dn_attn_mask_type_list = ['match2dn', 'dn2dn', 'group2group']
+dn_labelbook_size = 100
+
+match_unstable_error = False
+
+# for ema
+use_ema = True
+ema_decay = 0.9997
+ema_epoch = 0
+
+cls_no_bias = False
+num_body_points = 17  # for coco
+num_hand_points = 6 # for coco
+num_face_points = 6  # for coco
+num_group = 100
+num_box_decoder_layers = 2
+num_hand_face_decoder_layers = 4
+no_mmpose_keypoint_evaluator = True
+strong_aug = False
+
+body_model_test=\
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+body_model_train = \
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+# will be update in exp
+exp_name = 'output/exp52/dataset_debug'
+
+
+end_epoch = 150
+train_batch_size = 32
+
+scheduler = 'step'
+step_size = 20
+gamma = 0.1
+
+# continue
+continue_train = True
+pretrained_model_path = '../output/train_gta_synbody_ft_20230410_132110/model_dump/snapshot_2.pth.tar'
+
+# dataset setting
+# dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+# trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+dataset_list = ['INFERENCE_demo']
+trainset_3d = []
+trainset_2d = []
+trainset_partition = {
+            }
+trainset_humandata = []
+testset = 'INFERENCE_demo'
+train_sizes=[480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800]
+train_max_size=1333
+test_sizes=[800]
+test_max_size=1333
+no_aug=False
+# model
+use_cache = True
+
+## UBody setting
+train_sample_interval = 10
+test_sample_interval = 100
+make_same_len = False
+
+## input, output size
+input_body_shape = (256, 192)
+output_hm_shape = (16, 16, 12)
+input_hand_shape = (256, 256)
+output_hand_hm_shape = (16, 16, 16)
+output_face_hm_shape = (8, 8, 8)
+input_face_shape = (192, 192)
+focal = (5000, 5000)  # virtual focal lengths
+princpt = (input_body_shape[1] / 2, input_body_shape[0] / 2
+           )  # virtual principal point position
+body_3d_size = 2
+hand_3d_size = 0.3
+face_3d_size = 0.3
+camera_3d_size = 2.5
+
+bbox_ratio = 1.2
+
+## directory
+output_dir, model_dir, vis_dir, log_dir, result_dir, code_dir = None, None, None, None, None, None
+
+agora_benchmark = 'na' # 'agora_model', 'test_only'
+
+# strategy 
+data_strategy = 'balance' # 'balance' need to define total_data_len
+total_data_len = 'auto'

config/aios_smplx_agora_val.py

@@ -0,0 +1,265 @@
+
+num_classes = 2
+lr = 1e-04
+param_dict_type = 'default'
+lr_backbone = 1e-05
+lr_backbone_names = ['backbone.0']
+lr_linear_proj_names = ['reference_points', 'sampling_offsets']
+lr_linear_proj_mult = 0.1
+ddetr_lr_param = False
+batch_size = 2
+weight_decay = 0.0001
+epochs = 200
+lr_drop = 11
+save_checkpoint_interval = 1
+clip_max_norm = 0.1
+onecyclelr = False
+multi_step_lr = True
+lr_drop_list = [30, 60]
+
+modelname = 'aios_smplx'
+frozen_weights = None
+backbone = 'resnet50'
+use_checkpoint = False
+
+dilation = False
+position_embedding = 'sine'
+pe_temperatureH = 20
+pe_temperatureW = 20
+return_interm_indices = [1, 2, 3]
+backbone_freeze_keywords = None
+enc_layers = 6
+dec_layers = 6
+pre_norm = False
+dim_feedforward = 2048
+hidden_dim = 256
+dropout = 0.0
+nheads = 8
+num_queries = 900
+query_dim = 4
+num_patterns = 0
+random_refpoints_xy = False
+fix_refpoints_hw = -1
+dec_layer_number = None
+num_feature_levels = 4
+enc_n_points = 4
+dec_n_points = 4
+dln_xy_noise = 0.2
+dln_hw_noise = 0.2
+two_stage_type = 'standard'
+two_stage_bbox_embed_share = False
+two_stage_class_embed_share = False
+two_stage_learn_wh = False
+two_stage_default_hw = 0.05
+two_stage_keep_all_tokens = False
+rm_detach = None
+num_select = 50
+transformer_activation = 'relu'
+batch_norm_type = 'FrozenBatchNorm2d'
+
+masks = False
+losses = ["smpl_pose", "smpl_beta", "smpl_expr",
+        "smpl_kp2d","smpl_kp3d","smpl_kp3d_ra",'labels', 'boxes', "keypoints"]
+# losses = ['labels', 'boxes', "keypoints"]
+aux_loss = True
+set_cost_class = 2.0
+set_cost_bbox = 5.0
+set_cost_giou = 2.0
+set_cost_keypoints = 10.0
+set_cost_kpvis = 0.0
+set_cost_oks = 4.0
+cls_loss_coef = 2.0
+# keypoints_loss_coef = 10.0
+
+smpl_pose_loss_root_coef = 10 * 0.1
+smpl_pose_loss_body_coef = 1 * 0.1
+smpl_pose_loss_lhand_coef = 1 * 0.1
+smpl_pose_loss_rhand_coef = 1 * 0.1
+smpl_pose_loss_jaw_coef = 1 * 0.1
+smpl_beta_loss_coef = 0.01
+smpl_expr_loss_coef = 0.01
+
+# smpl_kp3d_loss_coef = 10
+smpl_body_kp3d_loss_coef = 10.0 * 0.1
+smpl_face_kp3d_loss_coef = 1.0 * 0.1
+smpl_lhand_kp3d_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_loss_coef = 1 * 0.1
+
+# kp3d ra
+smpl_body_kp3d_ra_loss_coef = 10 * 0.1
+smpl_face_kp3d_ra_loss_coef = 1 * 0.1
+smpl_lhand_kp3d_ra_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_ra_loss_coef = 1 * 0.1
+
+
+# smpl_kp2d_ba_loss_coef = 1.0
+smpl_body_kp2d_loss_coef = 10.0 * 0.1
+smpl_lhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_rhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_face_kp2d_loss_coef = 1.0 * 0.1
+
+smpl_body_kp2d_ba_loss_coef = 0 * 0.1
+smpl_face_kp2d_ba_loss_coef = 0 * 0.1
+smpl_lhand_kp2d_ba_loss_coef = 0 * 0.1
+smpl_rhand_kp2d_ba_loss_coef = 0 * 0.1
+
+bbox_loss_coef = 5.0
+body_bbox_loss_coef = 5.0
+lhand_bbox_loss_coef = 5.0
+rhand_bbox_loss_coef = 5.0
+face_bbox_loss_coef = 5.0
+
+giou_loss_coef = 2.0
+body_giou_loss_coef = 2.0
+rhand_giou_loss_coef = 2.0
+lhand_giou_loss_coef = 2.0
+face_giou_loss_coef = 2.0
+
+keypoints_loss_coef = 10.0
+rhand_keypoints_loss_coef = 10.0
+lhand_keypoints_loss_coef = 10.0
+face_keypoints_loss_coef = 10.0
+       
+oks_loss_coef=4.0
+rhand_oks_loss_coef = 0.5
+lhand_oks_loss_coef = 0.5
+face_oks_loss_coef = 4.0
+
+
+enc_loss_coef = 1.0
+interm_loss_coef = 1.0
+no_interm_box_loss = False
+focal_alpha = 0.25
+rm_self_attn_layers = None
+indices_idx_list = [1, 2, 3, 4, 5, 6, 7]
+
+decoder_sa_type = 'sa'
+matcher_type = 'HungarianMatcher'
+decoder_module_seq = ['sa', 'ca', 'ffn']
+nms_iou_threshold = -1
+
+dec_pred_bbox_embed_share = False
+dec_pred_class_embed_share = False
+dec_pred_pose_embed_share = False
+body_only = True
+
+# for dn
+use_dn = True
+dn_number = 100
+dn_box_noise_scale = 0.4
+dn_label_noise_ratio = 0.5
+embed_init_tgt = False
+dn_label_coef = 0.3
+dn_bbox_coef = 0.5
+dn_batch_gt_fuse = False
+dn_attn_mask_type_list = ['match2dn', 'dn2dn', 'group2group']
+dn_labelbook_size = 100
+
+match_unstable_error = False
+
+# for ema
+use_ema = True
+ema_decay = 0.9997
+ema_epoch = 0
+
+cls_no_bias = False
+num_body_points = 17  # for coco
+num_hand_points = 6 # for coco
+num_face_points = 6  # for coco
+num_group = 100
+num_box_decoder_layers = 2
+num_hand_face_decoder_layers = 4
+no_mmpose_keypoint_evaluator = True
+strong_aug = False
+
+body_model_test=\
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+body_model_train = \
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+# will be update in exp
+exp_name = 'output/exp52/dataset_debug'
+
+
+end_epoch = 150
+train_batch_size = 32
+
+scheduler = 'step'
+step_size = 20
+gamma = 0.1
+
+# continue
+continue_train = True
+pretrained_model_path = '../output/train_gta_synbody_ft_20230410_132110/model_dump/snapshot_2.pth.tar'
+
+# dataset setting
+# dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+# trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_2d = []
+trainset_partition = {
+            'AGORA_MM': 0.4, 
+            'BEDLAM': 0.7,
+            'COCO_NA': 1,
+            
+            # 'EgoBody_Egocentric': 1,
+            # 'EgoBody_Kinect': 1.0,
+            }
+trainset_humandata = []
+testset = 'INFERENCE_AGORA'
+train_sizes=[480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800]
+train_max_size=1333
+test_sizes=[800]
+test_max_size=1333
+no_aug=False
+# model
+use_cache = True
+
+## UBody setting
+train_sample_interval = 10
+test_sample_interval = 100
+make_same_len = False
+
+## input, output size
+input_body_shape = (256, 192)
+output_hm_shape = (16, 16, 12)
+input_hand_shape = (256, 256)
+output_hand_hm_shape = (16, 16, 16)
+output_face_hm_shape = (8, 8, 8)
+input_face_shape = (192, 192)
+focal = (5000, 5000)  # virtual focal lengths
+princpt = (input_body_shape[1] / 2, input_body_shape[0] / 2
+           )  # virtual principal point position
+body_3d_size = 2
+hand_3d_size = 0.3
+face_3d_size = 0.3
+camera_3d_size = 2.5
+
+bbox_ratio = 1.2
+
+## directory
+output_dir, model_dir, vis_dir, log_dir, result_dir, code_dir = None, None, None, None, None, None
+
+agora_benchmark = 'na' # 'agora_model', 'test_only'
+
+# strategy 
+data_strategy = 'balance' # 'balance' need to define total_data_len
+total_data_len = 'auto'

config/aios_smplx_bedlam.py

@@ -0,0 +1,265 @@
+
+num_classes = 2
+lr = 0.0001*1.414/10
+param_dict_type = 'default'
+lr_backbone = 1e-05*1.414/10
+lr_backbone_names = ['backbone.0']
+lr_linear_proj_names = ['reference_points', 'sampling_offsets']
+lr_linear_proj_mult = 0.1
+ddetr_lr_param = False
+batch_size = 2
+weight_decay = 0.0001
+epochs = 200
+lr_drop = 11
+save_checkpoint_interval = 1
+clip_max_norm = 0.1
+onecyclelr = False
+multi_step_lr = True
+lr_drop_list = [30, 60]
+
+modelname = 'aios_smplx'
+frozen_weights = None
+backbone = 'resnet50'
+use_checkpoint = False
+
+dilation = False
+position_embedding = 'sine'
+pe_temperatureH = 20
+pe_temperatureW = 20
+return_interm_indices = [1, 2, 3]
+backbone_freeze_keywords = None
+enc_layers = 6
+dec_layers = 6
+pre_norm = False
+dim_feedforward = 2048
+hidden_dim = 256
+dropout = 0.0
+nheads = 8
+num_queries = 900
+query_dim = 4
+num_patterns = 0
+random_refpoints_xy = False
+fix_refpoints_hw = -1
+dec_layer_number = None
+num_feature_levels = 4
+enc_n_points = 4
+dec_n_points = 4
+dln_xy_noise = 0.2
+dln_hw_noise = 0.2
+two_stage_type = 'standard'
+two_stage_bbox_embed_share = False
+two_stage_class_embed_share = False
+two_stage_learn_wh = False
+two_stage_default_hw = 0.05
+two_stage_keep_all_tokens = False
+rm_detach = None
+num_select = 50
+transformer_activation = 'relu'
+batch_norm_type = 'FrozenBatchNorm2d'
+
+masks = False
+losses = ["smpl_pose", "smpl_beta", "smpl_expr",
+        "smpl_kp2d","smpl_kp3d","smpl_kp3d_ra",'labels', 'boxes', "keypoints"]
+# losses = ['labels', 'boxes', "keypoints"]
+aux_loss = True
+set_cost_class = 2.0
+set_cost_bbox = 5.0
+set_cost_giou = 2.0
+set_cost_keypoints = 10.0
+set_cost_kpvis = 0.0
+set_cost_oks = 4.0
+cls_loss_coef = 2.0
+# keypoints_loss_coef = 10.0
+
+smpl_pose_loss_root_coef = 10 * 0.1
+smpl_pose_loss_body_coef = 1 * 0.1
+smpl_pose_loss_lhand_coef = 1 * 0.1
+smpl_pose_loss_rhand_coef = 1 * 0.1
+smpl_pose_loss_jaw_coef = 1 * 0.1
+smpl_beta_loss_coef = 0.01
+smpl_expr_loss_coef = 0.01
+
+# smpl_kp3d_loss_coef = 10
+smpl_body_kp3d_loss_coef = 10.0 * 0.1
+smpl_face_kp3d_loss_coef = 1.0 * 0.1
+smpl_lhand_kp3d_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_loss_coef = 1 * 0.1
+
+# kp3d ra
+smpl_body_kp3d_ra_loss_coef = 10 * 0.1
+smpl_face_kp3d_ra_loss_coef = 1 * 0.1
+smpl_lhand_kp3d_ra_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_ra_loss_coef = 1 * 0.1
+
+
+# smpl_kp2d_ba_loss_coef = 1.0
+smpl_body_kp2d_loss_coef = 10.0 * 0.1
+smpl_lhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_rhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_face_kp2d_loss_coef = 1.0 * 0.1
+
+smpl_body_kp2d_ba_loss_coef = 0 * 0.1
+smpl_face_kp2d_ba_loss_coef = 0 * 0.1
+smpl_lhand_kp2d_ba_loss_coef = 0 * 0.1
+smpl_rhand_kp2d_ba_loss_coef = 0 * 0.1
+
+bbox_loss_coef = 5.0
+body_bbox_loss_coef = 5.0
+lhand_bbox_loss_coef = 5.0
+rhand_bbox_loss_coef = 5.0
+face_bbox_loss_coef = 5.0
+
+giou_loss_coef = 2.0
+body_giou_loss_coef = 2.0
+rhand_giou_loss_coef = 2.0
+lhand_giou_loss_coef = 2.0
+face_giou_loss_coef = 2.0
+
+keypoints_loss_coef = 10.0
+rhand_keypoints_loss_coef = 10.0
+lhand_keypoints_loss_coef = 10.0
+face_keypoints_loss_coef = 10.0
+       
+oks_loss_coef=4.0
+rhand_oks_loss_coef = 0.5
+lhand_oks_loss_coef = 0.5
+face_oks_loss_coef = 4.0
+
+
+enc_loss_coef = 1.0
+interm_loss_coef = 1.0
+no_interm_box_loss = False
+focal_alpha = 0.25
+rm_self_attn_layers = None
+indices_idx_list = [1, 2, 3, 4, 5, 6, 7]
+
+decoder_sa_type = 'sa'
+matcher_type = 'HungarianMatcher'
+decoder_module_seq = ['sa', 'ca', 'ffn']
+nms_iou_threshold = -1
+
+dec_pred_bbox_embed_share = False
+dec_pred_class_embed_share = False
+dec_pred_pose_embed_share = False
+body_only = True
+
+# for dn
+use_dn = True
+dn_number = 100
+dn_box_noise_scale = 0.4
+dn_label_noise_ratio = 0.5
+embed_init_tgt = False
+dn_label_coef = 0.3
+dn_bbox_coef = 0.5
+dn_batch_gt_fuse = False
+dn_attn_mask_type_list = ['match2dn', 'dn2dn', 'group2group']
+dn_labelbook_size = 100
+
+match_unstable_error = False
+
+# for ema
+use_ema = True
+ema_decay = 0.9997
+ema_epoch = 0
+
+cls_no_bias = False
+num_body_points = 17  # for coco
+num_hand_points = 6 # for coco
+num_face_points = 6  # for coco
+num_group = 100
+num_box_decoder_layers = 2
+num_hand_face_decoder_layers = 4
+no_mmpose_keypoint_evaluator = True
+strong_aug = False
+
+body_model_test=\
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+body_model_train = \
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+# will be update in exp
+exp_name = 'output/exp52/dataset_debug'
+
+
+end_epoch = 150
+train_batch_size = 32
+
+scheduler = 'step'
+step_size = 20
+gamma = 0.1
+
+# continue
+continue_train = True
+pretrained_model_path = '../output/train_gta_synbody_ft_20230410_132110/model_dump/snapshot_2.pth.tar'
+
+# dataset setting
+# dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+# trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_2d = []
+trainset_partition = {
+            'AGORA_MM': 0.4, 
+            'BEDLAM': 0.7,
+            'COCO_NA': 1,
+            
+            # 'EgoBody_Egocentric': 1,
+            # 'EgoBody_Kinect': 1.0,
+            }
+trainset_humandata = []
+testset = 'INFERENCE_BEDLAM'
+train_sizes=[480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800]
+train_max_size=1333
+test_sizes=[800]
+test_max_size=1333
+no_aug=False
+# model
+use_cache = True
+
+## UBody setting
+train_sample_interval = 10
+test_sample_interval = 100
+make_same_len = False
+
+## input, output size
+input_body_shape = (256, 192)
+output_hm_shape = (16, 16, 12)
+input_hand_shape = (256, 256)
+output_hand_hm_shape = (16, 16, 16)
+output_face_hm_shape = (8, 8, 8)
+input_face_shape = (192, 192)
+focal = (5000, 5000)  # virtual focal lengths
+princpt = (input_body_shape[1] / 2, input_body_shape[0] / 2
+           )  # virtual principal point position
+body_3d_size = 2
+hand_3d_size = 0.3
+face_3d_size = 0.3
+camera_3d_size = 2.5
+
+bbox_ratio = 1.2
+
+## directory
+output_dir, model_dir, vis_dir, log_dir, result_dir, code_dir = None, None, None, None, None, None
+
+agora_benchmark = 'na' # 'agora_model', 'test_only'
+
+# strategy 
+data_strategy = 'balance' # 'balance' need to define total_data_len
+total_data_len = 'auto'

config/aios_smplx_demo.py

@@ -0,0 +1,259 @@
+
+num_classes = 2
+lr = 0.0001*1.414/10
+param_dict_type = 'default'
+lr_backbone = 1e-05*1.414/10
+lr_backbone_names = ['backbone.0']
+lr_linear_proj_names = ['reference_points', 'sampling_offsets']
+lr_linear_proj_mult = 0.1
+ddetr_lr_param = False
+batch_size = 2
+weight_decay = 0.0001
+epochs = 200
+lr_drop = 11
+save_checkpoint_interval = 1
+clip_max_norm = 0.1
+onecyclelr = False
+multi_step_lr = True
+lr_drop_list = [30, 60]
+
+modelname = 'aios_smplx'
+frozen_weights = None
+backbone = 'resnet50'
+use_checkpoint = False
+
+dilation = False
+position_embedding = 'sine'
+pe_temperatureH = 20
+pe_temperatureW = 20
+return_interm_indices = [1, 2, 3]
+backbone_freeze_keywords = None
+enc_layers = 6
+dec_layers = 6
+pre_norm = False
+dim_feedforward = 2048
+hidden_dim = 256
+dropout = 0.0
+nheads = 8
+num_queries = 900
+query_dim = 4
+num_patterns = 0
+random_refpoints_xy = False
+fix_refpoints_hw = -1
+dec_layer_number = None
+num_feature_levels = 4
+enc_n_points = 4
+dec_n_points = 4
+dln_xy_noise = 0.2
+dln_hw_noise = 0.2
+two_stage_type = 'standard'
+two_stage_bbox_embed_share = False
+two_stage_class_embed_share = False
+two_stage_learn_wh = False
+two_stage_default_hw = 0.05
+two_stage_keep_all_tokens = False
+rm_detach = None
+num_select = 50
+transformer_activation = 'relu'
+batch_norm_type = 'FrozenBatchNorm2d'
+
+masks = False
+losses = ["smpl_pose", "smpl_beta", "smpl_expr",
+        "smpl_kp2d","smpl_kp3d","smpl_kp3d_ra",'labels', 'boxes', "keypoints"]
+# losses = ['labels', 'boxes', "keypoints"]
+aux_loss = True
+set_cost_class = 2.0
+set_cost_bbox = 5.0
+set_cost_giou = 2.0
+set_cost_keypoints = 10.0
+set_cost_kpvis = 0.0
+set_cost_oks = 4.0
+cls_loss_coef = 2.0
+# keypoints_loss_coef = 10.0
+
+smpl_pose_loss_root_coef = 10 * 0.1
+smpl_pose_loss_body_coef = 1 * 0.1
+smpl_pose_loss_lhand_coef = 1 * 0.1
+smpl_pose_loss_rhand_coef = 1 * 0.1
+smpl_pose_loss_jaw_coef = 1 * 0.1
+smpl_beta_loss_coef = 0.01
+smpl_expr_loss_coef = 0.01
+
+# smpl_kp3d_loss_coef = 10
+smpl_body_kp3d_loss_coef = 10.0 * 0.1
+smpl_face_kp3d_loss_coef = 1.0 * 0.1
+smpl_lhand_kp3d_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_loss_coef = 1 * 0.1
+
+# kp3d ra
+smpl_body_kp3d_ra_loss_coef = 10 * 0.1
+smpl_face_kp3d_ra_loss_coef = 1 * 0.1
+smpl_lhand_kp3d_ra_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_ra_loss_coef = 1 * 0.1
+
+
+# smpl_kp2d_ba_loss_coef = 1.0
+smpl_body_kp2d_loss_coef = 10.0 * 0.1
+smpl_lhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_rhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_face_kp2d_loss_coef = 1.0 * 0.1
+
+smpl_body_kp2d_ba_loss_coef = 0 * 0.1
+smpl_face_kp2d_ba_loss_coef = 0 * 0.1
+smpl_lhand_kp2d_ba_loss_coef = 0 * 0.1
+smpl_rhand_kp2d_ba_loss_coef = 0 * 0.1
+
+bbox_loss_coef = 5.0
+body_bbox_loss_coef = 5.0
+lhand_bbox_loss_coef = 5.0
+rhand_bbox_loss_coef = 5.0
+face_bbox_loss_coef = 5.0
+
+giou_loss_coef = 2.0
+body_giou_loss_coef = 2.0
+rhand_giou_loss_coef = 2.0
+lhand_giou_loss_coef = 2.0
+face_giou_loss_coef = 2.0
+
+keypoints_loss_coef = 10.0
+rhand_keypoints_loss_coef = 10.0
+lhand_keypoints_loss_coef = 10.0
+face_keypoints_loss_coef = 10.0
+       
+oks_loss_coef=4.0
+rhand_oks_loss_coef = 0.5
+lhand_oks_loss_coef = 0.5
+face_oks_loss_coef = 4.0
+
+
+enc_loss_coef = 1.0
+interm_loss_coef = 1.0
+no_interm_box_loss = False
+focal_alpha = 0.25
+rm_self_attn_layers = None
+indices_idx_list = [1, 2, 3, 4, 5, 6, 7]
+
+decoder_sa_type = 'sa'
+matcher_type = 'HungarianMatcher'
+decoder_module_seq = ['sa', 'ca', 'ffn']
+nms_iou_threshold = -1
+
+dec_pred_bbox_embed_share = False
+dec_pred_class_embed_share = False
+dec_pred_pose_embed_share = False
+body_only = True
+
+# for dn
+use_dn = True
+dn_number = 100
+dn_box_noise_scale = 0.4
+dn_label_noise_ratio = 0.5
+embed_init_tgt = False
+dn_label_coef = 0.3
+dn_bbox_coef = 0.5
+dn_batch_gt_fuse = False
+dn_attn_mask_type_list = ['match2dn', 'dn2dn', 'group2group']
+dn_labelbook_size = 100
+
+match_unstable_error = False
+
+# for ema
+use_ema = True
+ema_decay = 0.9997
+ema_epoch = 0
+
+cls_no_bias = False
+num_body_points = 17  # for coco
+num_hand_points = 6 # for coco
+num_face_points = 6  # for coco
+num_group = 100
+num_box_decoder_layers = 2
+num_hand_face_decoder_layers = 4
+no_mmpose_keypoint_evaluator = True
+strong_aug = False
+
+body_model_test=\
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+body_model_train = \
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+# will be update in exp
+exp_name = 'output/exp52/dataset_debug'
+
+
+end_epoch = 150
+train_batch_size = 32
+
+scheduler = 'step'
+step_size = 20
+gamma = 0.1
+
+# continue
+continue_train = True
+pretrained_model_path = '../output/train_gta_synbody_ft_20230410_132110/model_dump/snapshot_2.pth.tar'
+
+# dataset setting
+# dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+# trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+dataset_list = ['INFERENCE_demo']
+trainset_3d = []
+trainset_2d = []
+trainset_partition = {
+            }
+trainset_humandata = []
+testset = 'INFERENCE_demo'
+train_sizes=[480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800]
+train_max_size=1333
+test_sizes=[800]
+test_max_size=1333
+no_aug=False
+# model
+use_cache = True
+
+## UBody setting
+train_sample_interval = 10
+test_sample_interval = 100
+make_same_len = False
+
+## input, output size
+input_body_shape = (256, 192)
+output_hm_shape = (16, 16, 12)
+input_hand_shape = (256, 256)
+output_hand_hm_shape = (16, 16, 16)
+output_face_hm_shape = (8, 8, 8)
+input_face_shape = (192, 192)
+focal = (5000, 5000)  # virtual focal lengths
+princpt = (input_body_shape[1] / 2, input_body_shape[0] / 2
+           )  # virtual principal point position
+body_3d_size = 2
+hand_3d_size = 0.3
+face_3d_size = 0.3
+camera_3d_size = 2.5
+
+bbox_ratio = 1.2
+
+## directory
+output_dir, model_dir, vis_dir, log_dir, result_dir, code_dir = None, None, None, None, None, None
+
+agora_benchmark = 'na' # 'agora_model', 'test_only'
+
+# strategy 
+data_strategy = 'balance' # 'balance' need to define total_data_len
+total_data_len = 'auto'

config/aios_smplx_inference.py

@@ -0,0 +1,265 @@
+
+num_classes = 2
+lr = 0.0001*1.414/10
+param_dict_type = 'default'
+lr_backbone = 1e-05*1.414/10
+lr_backbone_names = ['backbone.0']
+lr_linear_proj_names = ['reference_points', 'sampling_offsets']
+lr_linear_proj_mult = 0.1
+ddetr_lr_param = False
+batch_size = 2
+weight_decay = 0.0001
+epochs = 200
+lr_drop = 11
+save_checkpoint_interval = 1
+clip_max_norm = 0.1
+onecyclelr = False
+multi_step_lr = True
+lr_drop_list = [30, 60]
+
+modelname = 'aios_smplx'
+frozen_weights = None
+backbone = 'resnet50'
+use_checkpoint = False
+
+dilation = False
+position_embedding = 'sine'
+pe_temperatureH = 20
+pe_temperatureW = 20
+return_interm_indices = [1, 2, 3]
+backbone_freeze_keywords = None
+enc_layers = 6
+dec_layers = 6
+pre_norm = False
+dim_feedforward = 2048
+hidden_dim = 256
+dropout = 0.0
+nheads = 8
+num_queries = 900
+query_dim = 4
+num_patterns = 0
+random_refpoints_xy = False
+fix_refpoints_hw = -1
+dec_layer_number = None
+num_feature_levels = 4
+enc_n_points = 4
+dec_n_points = 4
+dln_xy_noise = 0.2
+dln_hw_noise = 0.2
+two_stage_type = 'standard'
+two_stage_bbox_embed_share = False
+two_stage_class_embed_share = False
+two_stage_learn_wh = False
+two_stage_default_hw = 0.05
+two_stage_keep_all_tokens = False
+rm_detach = None
+num_select = 50
+transformer_activation = 'relu'
+batch_norm_type = 'FrozenBatchNorm2d'
+
+masks = False
+losses = ["smpl_pose", "smpl_beta", "smpl_expr",
+        "smpl_kp2d","smpl_kp3d","smpl_kp3d_ra",'labels', 'boxes', "keypoints"]
+# losses = ['labels', 'boxes', "keypoints"]
+aux_loss = True
+set_cost_class = 2.0
+set_cost_bbox = 5.0
+set_cost_giou = 2.0
+set_cost_keypoints = 10.0
+set_cost_kpvis = 0.0
+set_cost_oks = 4.0
+cls_loss_coef = 2.0
+# keypoints_loss_coef = 10.0
+
+smpl_pose_loss_root_coef = 10 * 0.1
+smpl_pose_loss_body_coef = 1 * 0.1
+smpl_pose_loss_lhand_coef = 1 * 0.1
+smpl_pose_loss_rhand_coef = 1 * 0.1
+smpl_pose_loss_jaw_coef = 1 * 0.1
+smpl_beta_loss_coef = 0.01
+smpl_expr_loss_coef = 0.01
+
+# smpl_kp3d_loss_coef = 10
+smpl_body_kp3d_loss_coef = 10.0 * 0.1
+smpl_face_kp3d_loss_coef = 1.0 * 0.1
+smpl_lhand_kp3d_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_loss_coef = 1 * 0.1
+
+# kp3d ra
+smpl_body_kp3d_ra_loss_coef = 10 * 0.1
+smpl_face_kp3d_ra_loss_coef = 1 * 0.1
+smpl_lhand_kp3d_ra_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_ra_loss_coef = 1 * 0.1
+
+
+# smpl_kp2d_ba_loss_coef = 1.0
+smpl_body_kp2d_loss_coef = 10.0 * 0.1
+smpl_lhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_rhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_face_kp2d_loss_coef = 1.0 * 0.1
+
+smpl_body_kp2d_ba_loss_coef = 0 * 0.1
+smpl_face_kp2d_ba_loss_coef = 0 * 0.1
+smpl_lhand_kp2d_ba_loss_coef = 0 * 0.1
+smpl_rhand_kp2d_ba_loss_coef = 0 * 0.1
+
+bbox_loss_coef = 5.0
+body_bbox_loss_coef = 5.0
+lhand_bbox_loss_coef = 5.0
+rhand_bbox_loss_coef = 5.0
+face_bbox_loss_coef = 5.0
+
+giou_loss_coef = 2.0
+body_giou_loss_coef = 2.0
+rhand_giou_loss_coef = 2.0
+lhand_giou_loss_coef = 2.0
+face_giou_loss_coef = 2.0
+
+keypoints_loss_coef = 10.0
+rhand_keypoints_loss_coef = 10.0
+lhand_keypoints_loss_coef = 10.0
+face_keypoints_loss_coef = 10.0
+       
+oks_loss_coef=4.0
+rhand_oks_loss_coef = 0.5
+lhand_oks_loss_coef = 0.5
+face_oks_loss_coef = 4.0
+
+
+enc_loss_coef = 1.0
+interm_loss_coef = 1.0
+no_interm_box_loss = False
+focal_alpha = 0.25
+rm_self_attn_layers = None
+indices_idx_list = [1, 2, 3, 4, 5, 6, 7]
+
+decoder_sa_type = 'sa'
+matcher_type = 'HungarianMatcher'
+decoder_module_seq = ['sa', 'ca', 'ffn']
+nms_iou_threshold = -1
+
+dec_pred_bbox_embed_share = False
+dec_pred_class_embed_share = False
+dec_pred_pose_embed_share = False
+body_only = True
+
+# for dn
+use_dn = True
+dn_number = 100
+dn_box_noise_scale = 0.4
+dn_label_noise_ratio = 0.5
+embed_init_tgt = False
+dn_label_coef = 0.3
+dn_bbox_coef = 0.5
+dn_batch_gt_fuse = False
+dn_attn_mask_type_list = ['match2dn', 'dn2dn', 'group2group']
+dn_labelbook_size = 100
+
+match_unstable_error = False
+
+# for ema
+use_ema = True
+ema_decay = 0.9997
+ema_epoch = 0
+
+cls_no_bias = False
+num_body_points = 17  # for coco
+num_hand_points = 6 # for coco
+num_face_points = 6  # for coco
+num_group = 100
+num_box_decoder_layers = 2
+num_hand_face_decoder_layers = 4
+no_mmpose_keypoint_evaluator = True
+strong_aug = False
+
+body_model_test=\
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+body_model_train = \
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+# will be update in exp
+exp_name = 'output/exp52/dataset_debug'
+
+
+end_epoch = 150
+train_batch_size = 32
+
+scheduler = 'step'
+step_size = 20
+gamma = 0.1
+
+# continue
+continue_train = True
+pretrained_model_path = '../output/train_gta_synbody_ft_20230410_132110/model_dump/snapshot_2.pth.tar'
+
+# dataset setting
+# dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+# trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_2d = []
+trainset_partition = {
+            'AGORA_MM': 0.4, 
+            'BEDLAM': 0.7,
+            'COCO_NA': 1,
+            
+            # 'EgoBody_Egocentric': 1,
+            # 'EgoBody_Kinect': 1.0,
+            }
+trainset_humandata = []
+testset = 'INFERENCE'
+train_sizes=[480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800]
+train_max_size=1333
+test_sizes=[800]
+test_max_size=1333
+no_aug=False
+# model
+use_cache = True
+
+## UBody setting
+train_sample_interval = 10
+test_sample_interval = 100
+make_same_len = False
+
+## input, output size
+input_body_shape = (256, 192)
+output_hm_shape = (16, 16, 12)
+input_hand_shape = (256, 256)
+output_hand_hm_shape = (16, 16, 16)
+output_face_hm_shape = (8, 8, 8)
+input_face_shape = (192, 192)
+focal = (5000, 5000)  # virtual focal lengths
+princpt = (input_body_shape[1] / 2, input_body_shape[0] / 2
+           )  # virtual principal point position
+body_3d_size = 2
+hand_3d_size = 0.3
+face_3d_size = 0.3
+camera_3d_size = 2.5
+
+bbox_ratio = 1.2
+
+## directory
+output_dir, model_dir, vis_dir, log_dir, result_dir, code_dir = None, None, None, None, None, None
+
+agora_benchmark = 'na' # 'agora_model', 'test_only'
+
+# strategy 
+data_strategy = 'balance' # 'balance' need to define total_data_len
+total_data_len = 'auto'

config/aios_smplx_pretrain.py

@@ -0,0 +1,264 @@
+num_classes = 2
+lr = 0.0001
+param_dict_type = 'default'
+lr_backbone = 1e-05
+lr_backbone_names = ['backbone.0']
+lr_linear_proj_names = ['reference_points', 'sampling_offsets']
+lr_linear_proj_mult = 0.1
+ddetr_lr_param = False
+batch_size = 2
+weight_decay = 0.0001
+epochs = 200
+lr_drop = 11
+save_checkpoint_interval = 1
+clip_max_norm = 0.1
+onecyclelr = False
+multi_step_lr = True
+lr_drop_list = [30, 60]
+
+modelname = 'aios_smplx'
+frozen_weights = None
+backbone = 'resnet50'
+use_checkpoint = False
+
+dilation = False
+position_embedding = 'sine'
+pe_temperatureH = 20
+pe_temperatureW = 20
+return_interm_indices = [1, 2, 3]
+backbone_freeze_keywords = None
+enc_layers = 6
+dec_layers = 6
+pre_norm = False
+dim_feedforward = 2048
+hidden_dim = 256
+dropout = 0.0
+nheads = 8
+num_queries = 900
+query_dim = 4
+num_patterns = 0
+random_refpoints_xy = False
+fix_refpoints_hw = -1
+dec_layer_number = None
+num_feature_levels = 4
+enc_n_points = 4
+dec_n_points = 4
+dln_xy_noise = 0.2
+dln_hw_noise = 0.2
+two_stage_type = 'standard'
+two_stage_bbox_embed_share = False
+two_stage_class_embed_share = False
+two_stage_learn_wh = False
+two_stage_default_hw = 0.05
+two_stage_keep_all_tokens = False
+rm_detach = None
+num_select = 50
+transformer_activation = 'relu'
+batch_norm_type = 'FrozenBatchNorm2d'
+
+masks = False
+losses = ["smpl_pose", "smpl_beta", "smpl_expr",
+        "smpl_kp2d","smpl_kp3d","smpl_kp3d_ra",'labels', 'boxes', "keypoints"]
+# losses = ['labels', 'boxes', "keypoints"]
+aux_loss = True
+set_cost_class = 2.0
+set_cost_bbox = 5.0
+set_cost_giou = 2.0
+set_cost_keypoints = 10.0
+set_cost_kpvis = 0.0
+set_cost_oks = 4.0
+cls_loss_coef = 2.0
+# keypoints_loss_coef = 10.0
+
+smpl_pose_loss_root_coef = 10 * 0.1
+smpl_pose_loss_body_coef = 1 * 0.1
+smpl_pose_loss_lhand_coef = 1 * 0.1
+smpl_pose_loss_rhand_coef = 1 * 0.1
+smpl_pose_loss_jaw_coef = 1 * 0.1
+smpl_beta_loss_coef = 0.01
+smpl_expr_loss_coef = 0.01
+
+# smpl_kp3d_loss_coef = 10
+smpl_body_kp3d_loss_coef = 10.0 * 0.1
+smpl_face_kp3d_loss_coef = 1.0 * 0.1
+smpl_lhand_kp3d_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_loss_coef = 1 * 0.1
+
+# kp3d ra
+smpl_body_kp3d_ra_loss_coef = 10 * 0.1
+smpl_face_kp3d_ra_loss_coef = 1 * 0.1
+smpl_lhand_kp3d_ra_loss_coef = 1 * 0.1
+smpl_rhand_kp3d_ra_loss_coef = 1 * 0.1
+
+
+# smpl_kp2d_ba_loss_coef = 1.0
+smpl_body_kp2d_loss_coef = 10.0 * 0.1
+smpl_lhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_rhand_kp2d_loss_coef = 5.0 * 0.1
+smpl_face_kp2d_loss_coef = 1.0 * 0.1
+
+smpl_body_kp2d_ba_loss_coef = 0 * 0.1
+smpl_face_kp2d_ba_loss_coef = 0 * 0.1
+smpl_lhand_kp2d_ba_loss_coef = 0 * 0.1
+smpl_rhand_kp2d_ba_loss_coef = 0 * 0.1
+
+bbox_loss_coef = 5.0
+body_bbox_loss_coef = 5.0
+lhand_bbox_loss_coef = 5.0
+rhand_bbox_loss_coef = 5.0
+face_bbox_loss_coef = 5.0
+
+giou_loss_coef = 2.0
+body_giou_loss_coef = 2.0
+rhand_giou_loss_coef = 2.0
+lhand_giou_loss_coef = 2.0
+face_giou_loss_coef = 2.0
+
+keypoints_loss_coef = 10.0
+rhand_keypoints_loss_coef = 10.0
+lhand_keypoints_loss_coef = 10.0
+face_keypoints_loss_coef = 10.0
+       
+oks_loss_coef=4.0
+rhand_oks_loss_coef = 0.5
+lhand_oks_loss_coef = 0.5
+face_oks_loss_coef = 4.0
+
+
+enc_loss_coef = 1.0
+interm_loss_coef = 1.0
+no_interm_box_loss = False
+focal_alpha = 0.25
+rm_self_attn_layers = None
+indices_idx_list = [1, 2, 3, 4, 5, 6, 7]
+
+decoder_sa_type = 'sa'
+matcher_type = 'HungarianMatcher'
+decoder_module_seq = ['sa', 'ca', 'ffn']
+nms_iou_threshold = -1
+
+dec_pred_bbox_embed_share = False
+dec_pred_class_embed_share = False
+dec_pred_pose_embed_share = False
+body_only = True
+
+# for dn
+use_dn = True
+dn_number = 100
+dn_box_noise_scale = 0.4
+dn_label_noise_ratio = 0.5
+embed_init_tgt = False
+dn_label_coef = 0.3
+dn_bbox_coef = 0.5
+dn_batch_gt_fuse = False
+dn_attn_mask_type_list = ['match2dn', 'dn2dn', 'group2group']
+dn_labelbook_size = 100
+
+match_unstable_error = False
+
+# for ema
+use_ema = True
+ema_decay = 0.9997
+ema_epoch = 0
+
+cls_no_bias = False
+num_body_points = 17  # for coco
+num_hand_points = 6 # for coco
+num_face_points = 6  # for coco
+num_group = 100
+num_box_decoder_layers = 2
+num_hand_face_decoder_layers = 4
+no_mmpose_keypoint_evaluator = True
+strong_aug = False
+
+body_model_test=\
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+body_model_train = \
+    dict(
+        type='smplx',
+        keypoint_src='smplx',
+        num_expression_coeffs=10,
+        num_betas=10,
+        keypoint_dst='smplx_137',
+        model_path='data/body_models/smplx',
+        use_pca=False,
+        use_face_contour=True)
+
+# will be update in exp
+exp_name = 'output/exp52/dataset_debug'
+
+
+end_epoch = 150
+train_batch_size = 32
+
+scheduler = 'step'
+step_size = 20
+gamma = 0.1
+
+# continue
+continue_train = True
+pretrained_model_path = '../output/train_gta_synbody_ft_20230410_132110/model_dump/snapshot_2.pth.tar'
+
+# dataset setting
+# dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+# trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+dataset_list = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_3d = ['AGORA_MM','BEDLAM', 'COCO_NA']
+trainset_2d = []
+trainset_partition = {
+            'AGORA_MM': 0.4, 
+            'BEDLAM': 0.7,
+            'COCO_NA': 1,
+            
+            # 'EgoBody_Egocentric': 1,
+            # 'EgoBody_Kinect': 1.0,
+            }
+trainset_humandata = []
+testset = 'AGORA_MM'
+train_sizes=[480, 512, 544, 576, 608, 640, 672, 704, 736, 768, 800]
+train_max_size=1333
+test_sizes=[800]
+test_max_size=1333
+no_aug=False
+# model
+use_cache = True
+
+## UBody setting
+train_sample_interval = 10
+test_sample_interval = 100
+make_same_len = False
+
+## input, output size
+input_body_shape = (256, 192)
+output_hm_shape = (16, 16, 12)
+input_hand_shape = (256, 256)
+output_hand_hm_shape = (16, 16, 16)
+output_face_hm_shape = (8, 8, 8)
+input_face_shape = (192, 192)
+focal = (5000, 5000)  # virtual focal lengths
+princpt = (input_body_shape[1] / 2, input_body_shape[0] / 2
+           )  # virtual principal point position
+body_3d_size = 2
+hand_3d_size = 0.3
+face_3d_size = 0.3
+camera_3d_size = 2.5
+
+bbox_ratio = 1.2
+
+## directory
+output_dir, model_dir, vis_dir, log_dir, result_dir, code_dir = None, None, None, None, None, None
+
+agora_benchmark = 'na' # 'agora_model', 'test_only'
+
+# strategy 
+data_strategy = 'balance' # 'balance' need to define total_data_len
+total_data_len = 'auto'

config/config.py

@@ -0,0 +1,91 @@
+import os
+import os.path as osp
+import sys
+import datetime
+from mmcv import Config as MMConfig
+
+class Config(MMConfig):
+    def __init__(self, cfg_dict=None, cfg_text=None, filename=None):
+        super().__init__(cfg_dict, cfg_text, filename)
+
+    def get_config_fromfile(self, config_path):
+        self.config_path = config_path
+
+        cfg, _ = MMConfig._file2dict(self.config_path)
+        
+        self.merge_from_dict(cfg)
+        # #import ipdb;ipdb.set_trace()
+        # self.__dict__.update(dict(cfg))
+        # # update dir
+        dir_dict = {}
+        exp_name = 'exps62'
+        time_str = datetime.datetime.now().strftime('%Y%m%d_%H%M%S')
+        dir_dict['cur_dir'] = osp.dirname(os.path.abspath(__file__))
+        dir_dict['root_dir'] = osp.join(dir_dict['cur_dir'], '..')
+        dir_dict['output_dir'] = osp.join(dir_dict['root_dir'], exp_name)
+        dir_dict['result_dir'] = osp.join(dir_dict['output_dir'], 'result')
+        dir_dict['data_dir'] = osp.join(dir_dict['root_dir'], 'dataset')
+        dir_dict['human_model_path'] = osp.join('data/body_models')
+        self.merge_from_dict(dir_dict)
+        # 
+        # ## add some paths to the system root dir
+        sys.path.insert(0, osp.join(self.root_dir, 'common'))
+        sys.path.insert(0, osp.join(self.root_dir, 'united-perception_utils'))
+        sys.path.insert(0, osp.join(self.cur_dir, 'humanbench_utils'))
+        sys.path.insert(0, osp.join(self.cur_dir, 'dinov2_utils'))
+        sys.path.insert(0, osp.join(self.cur_dir, 'lora_utils'))
+        sys.path.insert(0, osp.join(self.cur_dir, 'vit_adapter_utils'))
+        from util.dir import add_pypath
+        # add_pypath(osp.join(self.data_dir))
+        for dataset in os.listdir('datasets'):
+            if dataset not in ['humandata.py', '__pycache__', 'dataset.py']:
+                add_pypath(osp.join(self.root_dir, 'data', dataset))
+        add_pypath('datasets')
+        add_pypath(self.data_dir)
+
+    def prepare_dirs(self, exp_name):
+        time_str = datetime.datetime.now().strftime('%Y%m%d_%H%M%S')
+        self.output_dir = osp.join(self.root_dir, f'{exp_name}_{time_str}')
+        self.model_dir = osp.join(self.output_dir, 'model_dump')
+        self.vis_dir = osp.join(self.output_dir, 'vis')
+        self.log_dir = osp.join(self.output_dir, 'log')
+        self.code_dir = osp.join(self.output_dir, 'code')
+        self.result_dir = osp.join(self.output_dir.split('/')[:-1])
+        from util.dir import make_folder
+        make_folder(self.model_dir)
+        make_folder(self.vis_dir)
+        make_folder(self.log_dir)
+        make_folder(self.code_dir)
+        make_folder(self.result_dir)
+
+        ## copy some code to log dir as a backup
+        copy_files = [
+            'main/train.py', 'main/test.py', 'common/base.py', 'main/OSX.py',
+            'common/nets', 'main/OSX_WoDecoder.py', 'data/dataset.py',
+            'data/MSCOCO/MSCOCO.py', 'data/AGORA/AGORA.py'
+        ]
+        for file in copy_files:
+            os.system(f'cp -r {self.root_dir}/{file} {self.code_dir}')
+
+    def update_test_config(self, testset, agora_benchmark, shapy_eval_split,
+                           pretrained_model_path, use_cache):
+        self.testset = testset
+        self.agora_benchmark = agora_benchmark
+        self.pretrained_model_path = pretrained_model_path
+        self.shapy_eval_split = shapy_eval_split
+        self.use_cache = use_cache
+
+    def update_config(self, num_gpus, exp_name):
+        self.num_gpus = num_gpus
+        self.exp_name = exp_name
+
+        self.prepare_dirs(self.exp_name)
+
+        # Save
+        cfg_save = MMConfig(self.__dict__)
+        cfg_save.dump(osp.join(self.code_dir, 'config_base.py'))
+
+
+cfg = Config()
+cfg.get_config_fromfile('config/aios_smplx.py')
+

data/body_models/J_regressor_extra.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cc968ea4f9855571e82f90203280836b01f13ee42a8e1b89d8d580b801242a89
+size 496160

data/body_models/J_regressor_h36m.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c655cd7013d7829eb9acbebf0e43f952a3fa0305a53c35880e39192bfb6444a0
+size 937168

data/body_models/J_regressor_mano_LEFT.txt

@@ -0,0 +1,1902 @@
+# 21 778
+0 4 0.0019103600293901542
+0 5 0.0027920646583394562
+0 6 0.00029390154298310065
+0 7 0.00014695077149155033
+0 25 0.0016164584864070536
+0 26 0.000440852314474651
+0 32 0.011756061719324026
+0 33 0.021234386480529024
+0 34 0.019838354151359296
+0 35 0.016311535635562088
+0 36 0.015870683321087434
+0 37 0.02343864805290228
+0 38 0.01671565025716385
+0 39 0.020499632623071272
+0 40 0.005437178545187362
+0 41 0.010139603232916973
+0 42 0.002645113886847906
+0 43 0.00014695077149155033
+0 44 0.02005878030859662
+0 45 0.02233651726671565
+0 50 0.01763409257898604
+0 51 0.01704628949301984
+0 52 0.019838354151359296
+0 53 0.02079353416605437
+0 54 0.00822924320352682
+0 55 0.00822924320352682
+0 78 0.011572373254959589
+0 79 0.011939750183688464
+0 84 0.01704628949301984
+0 85 0.019691403379867745
+0 88 0.005437178545187362
+0 89 0.0007347538574577516
+0 90 0.014548126377663484
+0 91 0.018736223365172666
+0 92 0.011645848640705364
+0 106 0.018515797207935343
+0 107 0.02204261572373255
+0 108 0.012417340191036004
+0 109 0.009992652461425423
+0 110 0.016311535635562088
+0 111 0.01880969875091844
+0 112 0.0073475385745775165
+0 113 0.0014695077149155032
+0 114 0.005731080088170463
+0 116 0.02204261572373255
+0 117 0.012123438648052902
+0 118 0.013005143277002204
+0 119 0.016385011021307863
+0 120 0.008155767817781044
+0 121 0.011315209404849376
+0 122 0.009037472446730345
+0 130 0.0073475385745775165
+0 131 0.00911094783247612
+0 178 0.001763409257898604
+0 179 0.002351212343864805
+0 190 0.019544452608376194
+0 191 0.019691403379867745
+0 192 0.01704628949301984
+0 193 0.016605437178545186
+0 200 0.002351212343864805
+0 203 0.00822924320352682
+0 204 0.007641440117560617
+0 205 0.01704628949301984
+0 207 0.001763409257898604
+0 208 0.005290227773695812
+0 209 0.01763409257898604
+0 210 0.019691403379867745
+0 211 0.019691403379867745
+0 214 0.011315209404849376
+0 215 0.011315209404849376
+0 216 0.007641440117560617
+0 217 0.00822924320352682
+0 218 0.002351212343864805
+0 219 0.0011756061719324026
+0 227 0.002351212343864805
+0 229 0.007788390889052168
+0 231 0.002204261572373255
+0 232 0.016311535635562088
+0 233 0.006759735488611315
+0 234 0.011168258633357825
+0 235 0.019544452608376194
+0 236 0.0016164584864070536
+0 239 0.011315209404849376
+0 241 0.0007347538574577516
+0 242 0.002351212343864805
+0 243 0.0036737692872887582
+0 244 0.0011756061719324026
+0 254 0.0064658339456282144
+0 255 0.0038207200587803084
+0 256 0.002351212343864805
+0 257 0.002351212343864805
+0 264 0.014107274063188832
+0 265 0.00440852314474651
+0 279 0.011315209404849376
+0 284 0.00896399706098457
+0 285 0.0029390154298310064
+1 0 0.014595751184471957
+1 1 0.025294207550053488
+1 2 0.019180803912578332
+1 3 0.01039278618370778
+1 4 0.03156044627846554
+1 5 0.025752712822864135
+1 6 0.014977838911814154
+1 7 0.023307351367874065
+1 8 0.005654898364664528
+1 9 0.009170105456212748
+1 10 0.002063273727647868
+1 11 0.0006113403637475165
+1 12 0.0018340210912425497
+1 14 0.001222680727495033
+1 15 7.641754546843957e-05
+1 16 0.0011462631820265935
+1 17 0.0004585052728106374
+1 18 0.00015283509093687913
+1 19 0.0003820877273421978
+1 22 7.641754546843957e-05
+1 24 0.01413724591166132
+1 25 0.019257221458046772
+1 26 0.024377197004432218
+1 27 0.017346782821335782
+1 28 0.0007641754546843956
+1 29 0.0022161088185847473
+1 30 0.0006877579092159561
+1 31 0.0005349228182790769
+1 32 0.0005349228182790768
+1 33 0.0005349228182790769
+1 34 0.0024071526822558465
+1 35 0.002445361454990066
+1 36 0.029802842732691428
+1 37 0.022122879413113253
+1 38 0.010029802842732692
+1 39 0.02334556014060829
+1 40 0.029344337459880795
+1 41 0.032171786642213054
+1 42 0.02009781445819961
+1 43 0.009934280910897143
+1 60 0.004355800091701055
+1 61 0.00855876509246523
+1 62 0.0004585052728106374
+1 63 0.003285954455142901
+1 64 0.0012990982729634726
+1 65 7.641754546843957e-05
+1 66 0.0019868561821794286
+1 67 0.004814305364511693
+1 68 0.008253094910591475
+1 69 0.0018340210912425497
+1 70 0.0003820877273421978
+1 71 7.641754546843957e-05
+1 88 0.021320495185694635
+1 89 0.013907993275256002
+1 90 0.01986856182179429
+1 91 0.013564114320648022
+1 92 0.003763564114320649
+1 93 0.0004585052728106374
+1 94 0.008329512456059913
+1 95 0.007565337001375517
+1 104 0.0027510316368638244
+1 105 0.0072596668195017595
+1 109 0.009705028274491823
+1 110 0.005654898364664528
+1 111 0.015436344184624792
+1 112 0.019180803912578332
+1 113 0.03339446736970809
+1 114 0.0340058077334556
+1 115 0.02559987773192725
+1 116 0.008405930001528351
+1 117 0.0017767079321412199
+1 118 0.00527281063732233
+1 119 0.00032477456824086816
+1 122 0.004967140455448571
+1 123 0.007259666819501758
+1 124 0.0016811860003056705
+1 125 0.0025217790004585057
+1 126 0.008176677365123033
+1 129 0.00030567018187375826
+1 145 0.00030567018187375826
+1 146 0.0006877579092159561
+1 147 7.641754546843957e-05
+1 152 7.641754546843957e-05
+1 157 0.002063273727647868
+1 158 0.0016047684548372307
+1 159 0.0032095369096744614
+1 188 0.0007641754546843956
+1 190 0.0019868561821794286
+1 191 0.0004585052728106374
+1 192 0.0016047684548372307
+1 193 0.005884151001069847
+1 207 0.00015283509093687913
+1 208 7.641754546843957e-05
+1 209 0.00030567018187375826
+1 216 0.0008405930001528353
+1 217 0.003897294818890417
+1 218 0.0008405930001528353
+1 219 0.0014519333639003516
+1 227 0.005502063273727648
+1 229 0.008635182637933671
+1 230 0.004126547455295736
+1 231 0.009705028274491824
+1 232 0.01245605991135565
+1 233 0.016888277548525142
+1 234 0.001413724591166132
+1 235 0.005654898364664528
+1 236 0.012838147638697846
+1 239 0.00026746140913953847
+1 240 0.01543634418462479
+1 241 0.0006877579092159561
+1 242 0.0032095369096744614
+1 248 0.004890722909980132
+1 249 0.0005349228182790769
+1 250 0.0015283509093687911
+1 251 0.0009170105456212748
+1 252 0.0029038667278007036
+1 253 0.005502063273727649
+1 254 0.0019868561821794286
+1 255 0.0002292526364053187
+1 264 0.028885832187070158
+1 265 0.029650007641754548
+1 266 0.006953996637628001
+1 267 0.002445361454990066
+1 268 0.00015283509093687913
+1 285 0.010087116001834023
+1 286 0.007794589637780836
+1 287 0.0025981965459269452
+1 697 0.0004585052728106374
+1 699 7.641754546843957e-05
+1 700 0.00030567018187375826
+1 704 0.0002292526364053187
+1 705 0.0008405930001528353
+1 706 7.641754546843957e-05
+2 0 0.0027531810402559712
+2 1 0.0034972840241089364
+2 2 0.007887491628841432
+2 3 0.0056551826772825355
+2 4 0.009152466701391472
+2 5 0.01674231713669172
+2 6 0.02708534861224793
+2 7 0.02209985862043307
+2 8 0.00833395341915321
+2 9 0.009152466701391472
+2 10 0.011682416846491553
+2 11 0.0055063620805119425
+2 12 0.005431951782126646
+2 13 0.0011161544757794478
+2 14 0.006176054765979612
+2 15 0.0017858471612471167
+2 16 0.0007441029838529652
+2 19 0.0003720514919264826
+2 26 0.000967333879008855
+2 27 0.0008929235806235583
+2 28 0.013245033112582783
+2 29 0.013765905201279856
+2 30 0.009970979983629735
+2 31 0.011384775652950369
+2 36 0.0023811295483294886
+2 37 0.00014882059677059304
+2 38 7.441029838529652e-05
+2 39 0.0020834883547883026
+2 40 0.0055063620805119425
+2 41 0.009896569685244438
+2 42 0.022843961604286034
+2 43 0.032666120991145166
+2 60 0.00364610462087953
+2 61 0.0017858471612471167
+2 62 0.0002976411935411861
+2 63 0.000967333879008855
+2 64 0.0014882059677059304
+2 65 0.0004464617903117792
+2 68 0.0002976411935411861
+2 69 7.441029838529652e-05
+2 88 0.01562616266091227
+2 89 0.027234169209018527
+2 90 0.00513431058858546
+2 91 0.0006696926854676687
+2 93 7.441029838529652e-05
+2 94 0.0005952823870823722
+2 104 0.025225091152615526
+2 105 0.017858471612471165
+2 113 0.0035716943224942334
+2 114 0.002604360443485378
+2 115 0.010566262370712107
+2 123 0.026787707418706754
+2 124 0.021504576233350697
+2 125 0.01882580549148002
+2 126 0.02083488354788303
+2 127 0.0002232308951558896
+2 128 0.0002976411935411861
+2 129 0.0017114368628618197
+2 144 0.0002232308951558896
+2 145 0.0013393853709353374
+2 158 0.002604360443485378
+2 193 0.0003720514919264826
+2 217 0.0007441029838529652
+2 219 0.0004464617903117792
+2 227 0.003199642830567751
+2 229 0.003125232532182454
+2 230 0.008854825507850286
+2 231 0.00982215938685914
+2 232 0.002009078056403006
+2 233 0.007813081330456134
+2 235 7.441029838529652e-05
+2 236 0.01912344668502121
+2 240 0.01480764937867401
+2 248 0.03318699307984225
+2 249 0.01823052310439765
+2 250 0.02887119577349505
+2 251 0.02500186025745963
+2 252 0.02864796487833916
+2 253 0.032889351886301064
+2 259 0.00014882059677059304
+2 264 0.0002232308951558896
+2 265 0.0005952823870823722
+2 266 0.015402931765756382
+2 267 0.01622144504799464
+2 286 0.02805268249125679
+2 287 0.025820373539697895
+2 697 0.014510008185132822
+2 698 0.008631594612694398
+2 699 0.011161544757794479
+2 700 0.01049185207232681
+2 701 0.00811072252399732
+2 702 0.013393853709353377
+2 703 0.010938313862638589
+2 704 0.008185132822382618
+2 705 0.02187662772527718
+2 706 0.018825805491480024
+2 707 0.011905647741647447
+2 708 0.007217798943373763
+2 709 0.005059900290200163
+2 710 0.003199642830567751
+2 711 0.0019346677580177095
+2 712 0.005952823870823722
+2 713 0.00364610462087953
+2 714 0.00364610462087953
+2 715 0.0026787707418706747
+2 716 0.0021578986531735995
+2 721 0.0006696926854676687
+2 722 0.0002232308951558896
+2 723 0.0002232308951558896
+2 725 0.0004464617903117792
+2 731 0.0032740531289530473
+2 732 0.0008185132822382618
+2 741 0.0005952823870823722
+2 742 0.0005208720886970756
+2 746 0.0002232308951558896
+2 749 0.0005208720886970756
+2 753 0.0034972840241089364
+2 754 0.004018156112806012
+2 755 0.0014882059677059304
+2 757 0.0008929235806235583
+2 758 0.0014137956693206339
+2 759 0.0003720514919264826
+2 760 7.441029838529652e-05
+3 6 0.0019164148301024542
+3 7 0.0014004569912287167
+3 8 0.000884499152354979
+3 9 0.00029483305078499295
+3 10 0.004422495761774894
+3 11 0.0011793322031399718
+3 12 0.0005896661015699859
+3 14 0.0011056239404437236
+3 28 0.011203655929829732
+3 29 0.0037591213975086604
+3 30 0.004496204024471142
+3 31 0.011645905506007222
+3 43 0.0019164148301024544
+3 89 0.0005896661015699859
+3 104 0.009729490675904768
+3 105 0.002137539618191199
+3 123 0.006412618854573597
+3 124 0.0187956069875433
+3 125 0.013414903810717178
+3 126 0.004938453600648632
+3 230 0.0007370826269624824
+3 231 0.00022112478808874474
+3 236 0.0005159578388737376
+3 240 0.0008844991523549787
+3 248 0.007665659320409817
+3 249 0.013120070759932186
+3 250 0.009434657625119773
+3 251 0.012088155082184712
+3 252 0.004348787499078646
+3 253 0.003022038770546178
+3 266 0.0029483305078499295
+3 267 0.0125304046583622
+3 286 0.002727205719761185
+3 287 0.005896661015699859
+3 697 0.01805852436058082
+3 698 0.019016731775632047
+3 699 0.021375396181911987
+3 700 0.01968010613989828
+3 701 0.023512935800103187
+3 702 0.01975381440259453
+3 703 0.021965062283481978
+3 704 0.019164148301024544
+3 705 0.015331318640819633
+3 706 0.017837399572492075
+3 707 0.02889363897692931
+3 708 0.02130168791921574
+3 709 0.027050932409523103
+3 710 0.024544851477850665
+3 711 0.0209331466057345
+3 712 0.0232181027493182
+3 713 0.023070686223925697
+3 714 0.024102601901673175
+3 715 0.018353357411365814
+3 716 0.017026608682833344
+3 717 0.0016952900420137097
+3 718 0.0062652023291811
+3 719 0.0033168718213311705
+3 720 0.00125304046583622
+3 721 0.016879192157440846
+3 722 0.01090882287904474
+3 723 0.008402741947372299
+3 724 0.004717328812559887
+3 725 0.010982531141740989
+3 726 0.0033168718213311705
+3 727 0.0008107908896587306
+3 730 7.370826269624824e-05
+3 731 0.022775853173140702
+3 732 0.018279649148669565
+3 733 0.009803198938601014
+3 734 0.003022038770546178
+3 735 0.0003685413134812412
+3 736 0.011719613768703471
+3 737 0.003906537922901157
+3 738 0.0008107908896587306
+3 739 0.013488612073413427
+3 740 0.005306994914129874
+3 741 0.021301687919215745
+3 742 0.019606397877202027
+3 743 0.0022112478808874476
+3 746 0.006338910591877348
+3 747 0.00125304046583622
+3 748 0.0016952900420137097
+3 749 0.009876907201297264
+3 750 0.003022038770546178
+3 751 7.370826269624824e-05
+3 753 0.025208225842116898
+3 754 0.0209331466057345
+3 755 0.023291811012014444
+3 756 0.017837399572492075
+3 757 0.021449104444608236
+3 758 0.01975381440259453
+3 759 0.01171961376870347
+3 760 0.01348861207341343
+3 761 0.003906537922901157
+3 762 0.005306994914129872
+3 763 0.007960492371194809
+3 764 0.0008107908896587306
+3 765 0.0003685413134812412
+3 767 0.0022112478808874476
+3 768 0.0011056239404437238
+4 745 1.0
+5 0 0.0012638674343491084
+5 1 0.0001404297149276787
+5 2 0.00035107428731919675
+5 3 0.002808594298553574
+5 8 0.004072461732902682
+5 9 0.0007723634321022329
+5 10 0.004774610307541076
+5 11 0.01418340120769555
+5 12 0.012357814913635726
+5 13 0.01930908580255582
+5 14 0.007934278893413846
+5 15 0.020011234377194213
+5 16 0.0021064457239151806
+5 17 0.0006319337171745541
+5 18 0.0022468754388428594
+5 19 0.009127931470299114
+5 21 0.00042128914478303613
+5 24 0.0009127931470299115
+5 25 7.021485746383936e-05
+5 26 0.0001404297149276787
+5 27 0.0010532228619575903
+5 28 0.0004212891447830361
+5 29 0.0015447268642044658
+5 30 0.003932032017975004
+5 31 0.0009127931470299115
+5 46 0.0006319337171745542
+5 47 0.00035107428731919675
+5 48 0.003721387445583485
+5 49 0.0027383794410897346
+5 56 0.0002808594298553574
+5 57 7.021485746383936e-05
+5 58 0.0010532228619575903
+5 59 0.0028788091560174134
+5 60 0.010040724617329027
+5 61 0.005687403454570988
+5 62 0.029981744137059403
+5 63 0.017483499508496
+5 64 0.02029209380704957
+5 65 0.024294340682488414
+5 66 0.0029490240134812527
+5 67 0.0011234377194214297
+5 68 0.005827833169498665
+5 69 0.00975986518747367
+5 74 0.00217666058137902
+5 75 0.0010532228619575903
+5 76 0.00035107428731919675
+5 77 0.00021064457239151807
+5 86 0.0007723634321022329
+5 87 0.0021064457239151806
+5 93 0.018536722370453586
+5 94 0.0016851565791321445
+5 95 0.0001404297149276787
+5 104 7.021485746383936e-05
+5 105 0.0001404297149276787
+5 127 0.023592192107850022
+5 128 0.02710293498104199
+5 129 0.020713382951832608
+5 132 0.023030473248139307
+5 133 0.005195899452324112
+5 134 0.005195899452324112
+5 135 0.01305996348827412
+5 136 0.008495997753124563
+5 137 0.014323830922623225
+5 138 0.01818564808313439
+5 139 0.011515236624069652
+5 140 0.008215138323269205
+5 143 0.010742873191967421
+5 144 0.016991995506249125
+5 145 0.010040724617329027
+5 146 0.00035107428731919675
+5 147 0.0011234377194214297
+5 149 0.013832326920376354
+5 150 0.016430276646538407
+5 151 0.010181154332256704
+5 152 0.011023732621822779
+5 155 0.00035107428731919675
+5 156 0.001966016008987502
+5 157 7.021485746383936e-05
+5 158 0.003932032017975004
+5 164 0.0034405280157281284
+5 165 0.005195899452324111
+5 166 0.0014745120067406266
+5 167 0.0014745120067406264
+5 168 0.026049712119084405
+5 169 0.02927959556242101
+5 170 0.023873051537705376
+5 171 0.016008987501755372
+5 172 0.027102934981041993
+5 173 0.016921780648785283
+5 174 0.005546973739643309
+5 175 0.005406544024715631
+5 176 0.013551467490520995
+5 177 0.00758320460609465
+5 183 7.021485746383936e-05
+5 185 0.009127931470299114
+5 186 0.017834573795815194
+5 187 0.008074708608341525
+5 189 0.007161915461311614
+5 194 0.010602443477039742
+5 195 0.01060244347703974
+5 206 0.0013340822918129478
+5 212 0.007091700603847775
+5 213 0.0013340822918129476
+5 219 0.0002808594298553574
+5 220 0.00435332116275804
+5 222 0.0002808594298553574
+5 223 0.00042128914478303613
+5 225 0.0016851565791321445
+5 226 0.00042128914478303613
+5 227 0.000983008004493751
+5 228 0.00975986518747367
+5 230 0.001825586294059823
+5 231 7.021485746383936e-05
+5 246 0.00035107428731919675
+5 258 0.020924027524224127
+5 259 0.022398539530964757
+5 260 0.015587698356972338
+5 261 0.012568459486027245
+5 262 0.009619435472545991
+5 263 0.01305996348827412
+5 266 0.0010532228619575903
+5 267 0.0005617188597107148
+5 268 0.004283106305294201
+5 269 0.0017553714365959837
+5 270 0.005266114309787951
+5 271 0.004844825165004915
+5 274 0.018045218368206713
+5 276 0.0002808594298553574
+5 277 0.00021064457239151807
+5 280 0.0001404297149276787
+5 288 0.00540654402471563
+5 290 7.021485746383936e-05
+5 358 0.0002808594298553574
+5 359 0.00035107428731919675
+5 362 0.00021064457239151807
+5 363 0.0002808594298553574
+5 365 7.021485746383936e-05
+5 366 0.0009127931470299116
+5 367 0.0013340822918129476
+5 368 0.005125684594860273
+5 369 0.0034405280157281284
+5 370 0.0013340822918129476
+5 371 0.00021064457239151807
+5 373 0.00042128914478303613
+5 375 0.00035107428731919675
+5 378 0.004493750877685719
+5 379 0.0034405280157281284
+5 380 0.004634180592613397
+5 383 0.00042128914478303613
+5 385 0.0016149417216683051
+5 386 0.001404297149276787
+5 387 0.0016851565791321445
+5 388 0.0002808594298553574
+5 399 0.0014745120067406264
+6 46 0.019904998869034157
+6 47 0.01960340797707909
+6 48 0.025559828093191583
+6 49 0.02352408957249491
+6 56 0.022166930558697125
+6 57 0.020131192038000453
+6 58 0.02194073738973083
+6 59 0.028952725627686037
+6 62 0.0005277840609213601
+6 65 0.00022619316896629722
+6 86 0.02382568046444997
+6 87 0.022543919173640955
+6 127 0.0012063635678202518
+6 128 0.0007539772298876573
+6 132 0.0006031817839101259
+6 133 0.017643067179371183
+6 134 0.02382568046444997
+6 135 0.01379778330694413
+6 136 0.01259141973912388
+6 137 0.004448465656337178
+6 138 0.003091306642539395
+6 139 0.009424715373595717
+6 140 0.012214431124180048
+6 143 0.0005277840609213601
+6 144 0.0012817612908090175
+6 150 0.0008293749528764231
+6 155 0.019678805700067855
+6 156 0.0244288622483601
+6 164 0.019980396592022914
+6 165 0.017944658071326246
+6 166 0.023222498680539848
+6 167 0.023901078187438737
+6 168 0.002789715750584332
+6 169 0.002186533966674206
+6 170 0.00987710171152831
+6 171 0.005881022393123726
+6 172 0.004071477041393349
+6 173 0.011837442509236221
+6 174 0.022166930558697128
+6 175 0.02382568046444997
+6 176 0.019377214808112796
+6 177 0.013119203800045236
+6 185 0.0016587499057528462
+6 186 0.004448465656337178
+6 187 0.0005277840609213601
+6 189 0.020809771544899342
+6 194 0.015154942320741913
+6 195 0.01839704440925884
+6 212 0.021262157882831936
+6 213 0.022317726004674656
+6 221 0.006333408731056322
+6 222 0.016210510442584633
+6 223 0.018472442132247607
+6 224 0.00987710171152831
+6 225 0.02744477116791073
+6 226 0.020583578375933047
+6 228 0.0005277840609213602
+6 237 0.012516022016135112
+6 238 0.011912840232224985
+6 245 0.011912840232224985
+6 258 0.0052024428862248355
+6 259 0.002337329412651738
+6 260 0.007162783683932745
+6 261 0.013043806077056472
+6 262 0.0016587499057528462
+6 263 0.007388976852899043
+6 272 0.014174771921887958
+6 273 0.012817612908090177
+6 274 0.0059564201161124925
+6 280 0.019301817085124028
+6 281 0.011385056171303627
+6 282 0.011460453894292393
+6 283 0.017643067179371186
+6 294 0.003920681595415819
+6 295 0.0069365905149664465
+6 296 0.0037698861494382865
+6 297 0.00512704516323607
+6 298 0.006634999623011385
+6 299 0.002789715750584332
+6 300 0.0021865339666742064
+6 301 0.0038452838724270517
+6 302 0.0005277840609213601
+6 303 0.0006031817839101259
+6 305 0.00030159089195506294
+6 316 0.0016587499057528462
+6 321 0.0009047726758651889
+6 330 0.0021111362436854408
+6 331 0.0015079544597753145
+6 340 0.00512704516323607
+6 341 0.004599261102314709
+6 342 0.0011309658448314859
+6 344 0.0007539772298876573
+6 345 0.00022619316896629722
+7 46 0.008690077640857611
+7 47 0.009188688653037966
+7 48 0.0033478167960680964
+7 49 0.0034902770852624832
+7 56 0.010898212123370611
+7 57 0.012322815015314481
+7 58 0.004202578531234419
+7 59 0.003276586651470902
+7 86 0.00648194315834461
+7 87 0.0016382933257354513
+7 133 0.00035615072298596765
+7 134 0.0015670631811382577
+7 155 0.009829759954412709
+7 156 0.004131348386637225
+7 164 0.0009259918797635161
+7 165 0.0006410713013747418
+7 166 0.003917657952845645
+7 167 0.0050573402664007405
+7 174 0.001638293325735451
+7 175 0.0014246028919438706
+7 189 0.0009259918797635161
+7 194 0.00028492057838877413
+7 195 0.0006410713013747418
+7 212 0.00042738086758316123
+7 213 0.0037039675190540643
+7 221 0.019517059619631027
+7 222 0.016739083980340477
+7 223 0.0143172590640359
+7 224 0.02443193959683738
+7 225 0.00683809388133058
+7 226 0.01111190255716219
+7 237 0.016739083980340477
+7 238 0.018092456727687157
+7 245 0.01367618776266116
+7 272 0.02236626540351877
+7 273 0.01923213904124225
+7 280 0.011040672412564997
+7 281 0.020086900776408578
+7 282 0.01859106773986751
+7 283 0.0165253935465489
+7 294 0.024004558729254222
+7 295 0.024075788873851416
+7 296 0.02443193959683738
+7 297 0.025357931476600898
+7 298 0.026283923356364414
+7 299 0.023933328584657028
+7 300 0.022722416126504736
+7 301 0.02514424104280932
+7 302 0.01738015528171522
+7 303 0.020941662511574897
+7 304 0.007835315905691288
+7 305 0.017380155281715225
+7 306 0.011396823135550965
+7 307 0.0036327373744568705
+7 308 0.0012821426027494836
+7 309 0.002777975639290548
+7 310 0.011966664292328516
+7 311 0.005342260844789515
+7 312 0.0038464278082484507
+7 313 0.0014958330365410642
+7 314 0.0007835315905691288
+7 315 0.008191466628677256
+7 316 0.022651185981907542
+7 317 0.00035615072298596765
+7 321 0.02101289265617209
+7 322 0.01225158487071729
+7 323 0.007764085761094094
+7 324 0.002564285205498967
+7 325 0.01994444048721419
+7 326 0.008690077640857611
+7 327 0.0024218249163045803
+7 328 0.0165253935465489
+7 329 0.006980554170524965
+7 330 0.028064676971294254
+7 331 0.021084122800769284
+7 332 0.0019232139041242254
+7 333 0.00021369043379158061
+7 334 0.010969442267967804
+7 335 0.0024930550609017737
+7 336 0.008690077640857611
+7 337 0.003988888097442838
+7 338 0.00028492057838877413
+7 340 0.019588289764228224
+7 341 0.0242182491630458
+7 342 0.021867654391338417
+7 343 0.014103568630244322
+7 344 0.018662297884464708
+7 345 0.014673409787021868
+7 346 0.006125792435358643
+7 347 0.009758529809815513
+7 348 0.0017095234703326447
+7 349 0.0031341263622765153
+7 350 0.004772419688011967
+7 351 0.0006410713013747418
+7 352 0.0008547617351663223
+7 353 0.00042738086758316123
+7 354 0.001068452168957903
+7 355 0.0009972220243607095
+8 317 1.0
+9 11 0.0002498906728306366
+9 13 0.0002498906728306366
+9 14 0.0009995626913225464
+9 15 0.0022490160554757294
+9 16 0.0029986880739676387
+9 17 0.002249016055475729
+9 18 0.007746610857749733
+9 19 0.00949584556756419
+9 20 0.0013743987005685012
+9 21 0.00437308677453614
+9 22 0.0009995626913225461
+9 23 0.00018741800462297744
+9 48 0.0004997813456612732
+9 59 0.0002498906728306366
+9 62 0.0014368713687761604
+9 63 0.000874617354907228
+9 64 6.247266820765915e-05
+9 65 6.247266820765915e-05
+9 66 0.0024989067283063657
+9 67 0.000437308677453614
+9 68 0.0006871993502842506
+9 69 0.0029986880739676387
+9 71 0.0004997813456612732
+9 74 0.015555694383707127
+9 75 0.017867183107390515
+9 76 0.017242456425313923
+9 77 0.00868370088086462
+9 83 6.247266820765915e-05
+9 87 0.0004997813456612732
+9 93 0.0033110514150059348
+9 127 0.0006247266820765914
+9 132 0.004810395451989753
+9 133 0.0006247266820765914
+9 135 0.0001249453364153183
+9 136 0.0004997813456612732
+9 137 0.015555694383707127
+9 138 0.007246829512088461
+9 139 0.005997376147935278
+9 140 0.008683700880864622
+9 141 0.005997376147935278
+9 142 0.0025613793965140247
+9 143 0.015743112388330104
+9 144 0.009558318235771848
+9 145 0.0032485787467982754
+9 146 0.0015618167051914785
+9 147 0.006122321484350596
+9 148 0.0025613793965140247
+9 149 0.0071843568438808006
+9 150 0.01243206097332417
+9 151 0.013993877678515648
+9 152 0.007809083525957393
+9 157 0.0001249453364153183
+9 158 0.0023114887236833884
+9 160 0.0019991253826450927
+9 161 0.0002498906728306366
+9 162 0.0005622540138689324
+9 163 0.0021240707190604106
+9 164 0.0029362154057599797
+9 165 0.002561379396514025
+9 166 0.0007496720184919098
+9 167 0.0007496720184919097
+9 168 0.002124070719060411
+9 169 0.0003123633410382957
+9 170 0.0006871993502842506
+9 171 0.002249016055475729
+9 174 0.0028737427375523207
+9 175 0.0018741800462297744
+9 176 0.009433372899356529
+9 177 0.006247266820765914
+9 181 0.00018741800462297744
+9 182 0.0009995626913225464
+9 183 0.004248141438120822
+9 185 0.019179109139751356
+9 186 0.01661772974323733
+9 187 0.019054163803336036
+9 194 0.0015618167051914785
+9 195 0.0001249453364153183
+9 196 0.0004997813456612732
+9 197 0.0014993440369838195
+9 198 0.0003748360092459549
+9 199 0.0001249453364153183
+9 202 6.247266820765915e-05
+9 206 0.013181732991816079
+9 207 6.247266820765915e-05
+9 212 0.0018741800462297742
+9 213 0.0002498906728306366
+9 218 0.0003123633410382957
+9 219 0.0006871993502842506
+9 220 0.014868495033422876
+9 225 0.0006247266820765914
+9 227 0.0006871993502842506
+9 228 0.021802961204473042
+9 230 0.0002498906728306366
+9 246 0.020803398513150495
+9 247 0.017304929093521583
+9 258 0.0004997813456612732
+9 259 0.0027487974011370024
+9 260 0.0017492347098144558
+9 261 0.002623852064721684
+9 262 0.01974136315362029
+9 263 0.01655525707502967
+9 268 0.007746610857749734
+9 269 0.02167801586805772
+9 270 0.019054163803336036
+9 271 0.011932279627662898
+9 274 0.0066221028300118695
+9 275 0.0007496720184919098
+9 276 0.016742675079652648
+9 277 0.02205285187730368
+9 288 0.022427687886549634
+9 289 0.0003123633410382957
+9 290 0.00730930218029612
+9 291 0.005685012806896982
+9 292 0.0057474854751046415
+9 293 0.008933591553695257
+9 356 0.0014993440369838195
+9 357 0.0014993440369838193
+9 358 0.00668457549821953
+9 359 0.004685450115574436
+9 360 0.0007496720184919098
+9 361 0.0007496720184919098
+9 362 0.0024989067283063657
+9 363 0.0038733054288748667
+9 364 0.0014368713687761604
+9 365 0.004498032110951459
+9 366 0.009933154245017804
+9 367 0.010245517586056099
+9 368 0.015993003061160742
+9 369 0.015993003061160742
+9 370 0.021115761854188793
+9 371 0.01693009308427563
+9 372 0.0009995626913225464
+9 373 0.0037483600924595483
+9 374 0.008996064221902918
+9 375 0.012432060973324168
+9 376 0.004498032110951458
+9 377 0.0031861060785906164
+9 378 0.017554819766352217
+9 379 0.01749234709814456
+9 380 0.01649278440682201
+9 381 0.008308864871618667
+9 382 0.006434684825388891
+9 383 0.016055475729368402
+9 384 0.012557006309739488
+9 385 0.01018304491784844
+9 386 0.015180858374461174
+9 387 0.01155744361841694
+9 388 0.009058536890110576
+9 389 0.0028112700693446614
+9 391 0.00018741800462297744
+9 392 0.0005622540138689324
+9 394 0.0018117073780221152
+9 395 0.0004997813456612732
+9 399 0.01611794839757606
+9 402 0.0008746173549072279
+9 470 0.0007496720184919098
+9 471 0.0004997813456612732
+9 478 0.0007496720184919098
+9 479 0.0004997813456612732
+9 480 0.0026863247329293434
+9 481 0.002623852064721684
+9 483 0.0001249453364153183
+9 484 0.0001249453364153183
+9 485 0.0014993440369838195
+9 486 0.0004997813456612732
+9 488 0.008996064221902916
+9 489 0.006059848816142937
+9 490 0.006497157493596552
+9 491 0.0001249453364153183
+9 492 0.0003748360092459549
+9 493 0.001311926032360842
+9 494 0.000437308677453614
+9 495 0.0017492347098144558
+9 496 0.002623852064721684
+9 497 0.0027487974011370024
+9 498 0.0006247266820765914
+9 509 0.0020615980508527517
+9 510 0.0003748360092459549
+9 579 0.0019991253826450927
+10 74 0.0005264345341054373
+10 75 0.0021809430698653833
+10 76 0.000752049334436339
+10 137 0.000827254267879973
+10 143 0.0006016394675490712
+10 150 0.0003008197337745356
+10 151 0.0006016394675490712
+10 185 0.004361886139730767
+10 186 0.0010528690682108748
+10 187 0.003910656539068963
+10 206 0.0001504098668872678
+10 220 0.0003008197337745356
+10 228 0.0030834022711889904
+10 246 0.003985861472512596
+10 247 0.0012784838685417762
+10 262 0.003910656539068963
+10 263 0.0011280740016545085
+10 269 0.0032338121380762574
+10 270 0.002857787470858088
+10 271 0.0003008197337745356
+10 276 0.000902459201323607
+10 277 0.00556516507482891
+10 288 0.0027825825374144545
+10 356 0.020305332029781156
+10 357 0.019703692562232082
+10 358 0.02549447243739189
+10 359 0.023764758968188315
+10 360 0.02587049710461006
+10 361 0.022486275099646538
+10 362 0.022411070166202904
+10 363 0.02278709483342107
+10 364 0.026321726705271865
+10 365 0.02007971722945025
+10 366 0.016093855756937656
+10 367 0.022260660299315636
+10 368 0.011882379484094157
+10 369 0.009400616680454237
+10 370 0.00962623148078514
+10 371 0.011431149883432353
+10 372 0.021583815898322933
+10 373 0.024742423102955553
+10 374 0.01947807776190118
+10 375 0.01789877415958487
+10 376 0.023388734300970146
+10 377 0.023689554034744677
+10 378 0.009400616680454237
+10 379 0.005865984808603443
+10 380 0.01135594494998872
+10 381 0.022486275099646538
+10 382 0.015341806422501316
+10 383 0.01135594494998872
+10 384 0.01158155975031962
+10 385 0.019703692562232082
+10 386 0.01504098668872678
+10 387 0.018124388959915774
+10 388 0.010077461081446944
+10 389 0.02293750470030834
+10 390 0.01383770775362864
+10 391 0.017372339625479433
+10 392 0.019703692562232086
+10 393 0.011882379484094157
+10 394 0.024667218169511923
+10 395 0.024667218169511916
+10 396 0.012333609084755958
+10 397 0.011506354816875987
+10 398 0.013236068286079568
+10 399 0.0070692637437015865
+10 400 0.01940287282845755
+10 401 0.016093855756937656
+10 402 0.020530946830112053
+10 403 0.008197337745356097
+10 404 0.01759795442581033
+10 405 0.021508610964879295
+10 406 0.008197337745356095
+10 407 0.013988117620515906
+10 408 0.008949387079792434
+10 409 0.006467624276152515
+10 410 0.005264345341054373
+10 411 0.005565165074828909
+10 412 0.003835451605625329
+10 413 0.002105738136421749
+10 414 0.0012784838685417764
+10 415 0.002556967737083553
+10 417 7.52049334436339e-05
+10 420 0.0020305332029781154
+10 421 0.0006016394675490712
+10 422 0.0006016394675490712
+10 427 7.52049334436339e-05
+10 430 0.004737910806948936
+10 431 0.002331352936752651
+10 432 0.0001504098668872678
+10 440 0.0010528690682108748
+10 441 0.0021057381364217496
+10 446 7.52049334436339e-05
+10 452 0.004512296006618034
+10 453 0.003609836805294428
+10 454 0.0006016394675490712
+10 456 0.0006016394675490712
+10 457 0.0004512296006618035
+11 356 0.011297349184080336
+11 357 0.011888060252528984
+11 358 0.004430333013364838
+11 359 0.004430333013364838
+11 360 0.009229860444510078
+11 361 0.011371188067636416
+11 362 0.0038396219449161927
+11 363 0.002805877575131064
+11 364 0.005759432917374288
+11 365 0.0014767776711216124
+11 366 0.0003691944177804031
+11 367 0.0014029387875655322
+11 372 0.011371188067636418
+11 373 0.004504171896920917
+11 374 0.0012552610204533705
+11 375 0.0011075832533412094
+11 376 0.005316399616037805
+11 377 0.005685594033818208
+11 381 0.001772133205345935
+11 382 0.0003691944177804031
+11 385 0.00118142213689729
+11 386 0.0005168721848925644
+11 387 0.0011075832533412094
+11 388 7.383888355608063e-05
+11 389 0.0031012331093553864
+11 390 0.019345787491693123
+11 391 0.010928154766299934
+11 392 0.01299564350587019
+11 393 0.02082256516281474
+11 394 0.0057594329173742895
+11 395 0.00945137709517832
+11 396 0.017352137635678947
+11 397 0.02001033744369785
+11 398 0.018238204238351912
+11 400 0.01794284870412759
+11 401 0.019124270841024884
+11 402 0.016170715498781657
+11 403 0.022816215018828915
+11 404 0.01727829875212287
+11 405 0.014546260060547885
+11 406 0.0239976371557262
+11 407 0.022963892785941076
+11 408 0.02695119249796943
+11 409 0.023776120505057962
+11 410 0.019493465258805284
+11 411 0.023849959388614037
+11 412 0.026581998080189025
+11 413 0.020601048512146496
+11 414 0.019493465258805288
+11 415 0.02163479288193162
+11 416 0.004873366314701322
+11 417 0.007900760540500627
+11 418 0.0042088163626965965
+11 419 0.0016982943217898545
+11 420 0.018238204238351912
+11 421 0.012035738019641142
+11 422 0.012331093553865465
+11 423 0.0055379162667060465
+11 424 0.004061138595584434
+11 425 0.0016982943217898542
+11 426 0.0008122277191168869
+11 427 0.00834379384183711
+11 428 0.0005168721848925643
+11 429 0.0015506165546776932
+11 430 0.023406926087277558
+11 431 0.019124270841024884
+11 432 0.016392232149449903
+11 433 0.005907110684486449
+11 434 0.0019198109724580966
+11 435 0.015432326663220851
+11 436 0.006940855054271579
+11 437 0.0013290999040094513
+11 438 0.013364837923650594
+11 439 0.00694085505427158
+11 440 0.02126559846415122
+11 441 0.02355460385438972
+11 442 0.002732038691574983
+11 444 7.383888355608063e-05
+11 446 0.010854315882743852
+11 447 0.0031012331093553864
+11 448 0.007753082773388465
+11 449 0.0018459720889020155
+11 450 0.00044303330133648377
+11 451 0.00044303330133648377
+11 452 0.023776120505057962
+11 453 0.02229934283393635
+11 454 0.02126559846415122
+11 455 0.013290999040094512
+11 456 0.018385882005464073
+11 457 0.015580004430333012
+11 458 0.010189765930739126
+11 459 0.012035738019641142
+11 460 0.0034704275271357893
+11 461 0.004578010780476998
+11 462 0.005907110684486449
+11 463 0.000590711068448645
+11 464 0.000590711068448645
+11 465 0.0002953555342243225
+11 466 0.0019936498560141768
+11 467 0.0013290999040094513
+12 445 1.0
+13 16 0.0014635288607891346
+13 17 0.002575810794988877
+13 18 0.005737033134293408
+13 19 0.001990399250673223
+13 20 0.007785973539398196
+13 21 0.008664090855871677
+13 22 0.002985598876009834
+13 23 0.002224563868399485
+13 63 5.854115443156538e-05
+13 66 0.0018147757873785268
+13 67 0.0006439526987472192
+13 68 0.0002927057721578269
+13 69 0.0008195761620419153
+13 70 0.0007024938531787846
+13 71 0.0033953869570307925
+13 72 0.0024001873316941806
+13 73 0.00023416461772626153
+13 74 0.009308043554618896
+13 75 0.007551808921671934
+13 76 0.01890879288139562
+13 77 0.013230300901533777
+13 80 0.0013464465519260039
+13 81 0.0002927057721578269
+13 82 0.0016976934785153963
+13 83 0.0040978808102095764
+13 93 0.00017562346329469617
+13 100 0.00017562346329469617
+13 102 0.00011708230886313077
+13 103 0.00035124692658939234
+13 137 0.00011708230886313077
+13 141 0.020021074815595362
+13 142 0.016625687858564567
+13 143 0.0016391523240838306
+13 144 0.0005268703898840885
+13 145 0.0002927057721578269
+13 146 0.002868516567146704
+13 147 0.006673691605198454
+13 148 0.008839714319166374
+13 149 0.0002927057721578269
+13 150 0.0002927057721578269
+13 151 0.0012293642430628731
+13 152 0.0011122819341997424
+13 157 0.0008781173164734808
+13 158 0.0004097880810209577
+13 160 0.02681184872965695
+13 161 0.023592085235920848
+13 162 0.03096827069429809
+13 163 0.02476290832455216
+13 178 0.0002927057721578269
+13 179 5.854115443156538e-05
+13 180 0.0009366584709050461
+13 181 0.00444912773679897
+13 182 0.013464465519260038
+13 183 0.0167427701674277
+13 184 0.00017562346329469617
+13 185 5.854115443156538e-05
+13 186 0.0002927057721578269
+13 187 0.0008195761620419153
+13 196 0.017503805175038047
+13 197 0.023416461772626154
+13 198 0.023416461772626154
+13 199 0.02921203606135113
+13 201 0.0018733169418100922
+13 202 0.006439526987472192
+13 206 0.015162158997775435
+13 207 0.0006439526987472192
+13 218 0.0007610350076103501
+13 219 0.00046832923545252306
+13 220 0.006673691605198454
+13 227 0.00011708230886313077
+13 228 0.0009951996253366115
+13 246 0.0106544901065449
+13 247 0.014576747453459781
+13 262 0.00011708230886313077
+13 268 0.0033368458025992264
+13 269 0.010420325488818641
+13 270 0.0035710104203254887
+13 271 0.002985598876009834
+13 275 0.009834913944502985
+13 276 0.02142606252195293
+13 277 0.01164968973188151
+13 278 0.00035124692658939234
+13 288 0.004741833508956796
+13 289 0.014693829762322912
+13 290 0.02207001522070015
+13 291 0.017913593256059006
+13 292 0.011005737033134292
+13 293 0.010478866643250203
+13 358 0.0003512469265893923
+13 363 5.854115443156538e-05
+13 365 0.00035124692658939234
+13 366 0.0007024938531787846
+13 367 0.00017562346329469617
+13 368 0.00017562346329469617
+13 369 0.0009951996253366115
+13 370 0.005151621589977754
+13 371 0.005385786207704015
+13 374 0.0016976934785153963
+13 375 0.0017562346329469615
+13 376 0.0004097880810209577
+13 377 0.0003512469265893923
+13 378 0.00046832923545252306
+13 379 0.0015220700152206996
+13 381 0.0014635288607891346
+13 382 0.0009951996253366115
+13 383 0.00532724505327245
+13 384 0.0037466338836201845
+13 386 0.0011708230886313077
+13 387 0.00011708230886313077
+13 388 0.0012293642430628731
+13 389 0.0002927057721578269
+13 394 0.00017562346329469617
+13 399 0.0033953869570307925
+13 468 5.854115443156538e-05
+13 469 0.0011122819341997424
+13 470 0.0027514342582835734
+13 471 0.0012879053974944384
+13 474 5.854115443156538e-05
+13 475 0.0002927057721578269
+13 476 0.0002927057721578269
+13 477 0.0018147757873785268
+13 478 0.0020489404051047887
+13 479 0.0011122819341997424
+13 480 0.004332045427935838
+13 481 0.006556609296335323
+13 483 0.00046832923545252306
+13 484 0.012352183585060298
+13 485 0.014869453225617611
+13 486 0.005912656597588104
+13 487 0.004214963119072708
+13 488 0.01164968973188151
+13 489 0.015806111696522657
+13 490 0.008312843929282283
+13 491 0.009834913944502985
+13 492 0.006146821215314366
+13 493 0.015513405924364829
+13 494 0.02007961597002693
+13 495 0.0024001873316941806
+13 496 0.008956796628029503
+13 497 0.004741833508956796
+13 498 0.003512469265893923
+13 499 0.002517269640557311
+13 501 0.0005854115443156538
+13 502 0.0004097880810209577
+13 504 0.001990399250673223
+13 505 0.00040978808102095764
+13 509 0.010478866643250205
+13 510 0.02207001522070015
+13 513 0.0012293642430628731
+13 579 0.021660227139679192
+13 580 0.0002927057721578269
+13 581 0.00011708230886313077
+13 582 0.0012879053974944388
+13 583 0.0018147757873785272
+13 584 5.854115443156538e-05
+13 585 0.00011708230886313077
+13 586 0.0011122819341997422
+13 587 0.0008195761620419154
+13 589 0.0007610350076103501
+13 590 0.003395386957030792
+13 591 0.0026928931038520073
+13 592 0.009834913944502985
+13 593 0.009834913944502985
+13 594 0.00011708230886313077
+13 595 0.0013464465519260039
+13 596 0.0015806111696522653
+13 597 0.0002927057721578269
+13 598 0.00023416461772626153
+13 599 0.0009951996253366115
+13 600 0.0002927057721578269
+13 601 0.0012293642430628731
+13 602 0.00046832923545252306
+13 603 0.00011708230886313077
+13 604 0.003980798501346446
+13 605 0.013523006673691603
+13 606 0.011591148577449948
+13 607 0.006263903524177495
+13 608 0.014693829762322912
+13 610 0.0003512469265893923
+13 611 0.0012293642430628734
+13 612 5.854115443156538e-05
+13 613 0.005327245053272449
+13 614 0.0019318580962416575
+13 615 0.006615150450766888
+13 616 0.0026928931038520073
+13 617 0.0002927057721578269
+13 627 0.005268703898840884
+13 630 0.00011708230886313077
+13 696 0.00023416461772626153
+13 769 0.00076103500761035
+13 770 0.004683292354525231
+13 771 0.0011122819341997424
+13 772 5.854115443156538e-05
+13 774 0.00076103500761035
+13 775 0.003512469265893923
+13 776 0.008020138157124457
+14 74 0.0005157677571470676
+14 75 0.0005157677571470676
+14 76 0.004273504273504274
+14 77 0.0008104921898025347
+14 141 0.002799882110226938
+14 142 0.0003684055408193339
+14 160 0.001326259946949602
+14 161 0.0005894488653109342
+14 162 0.004420866489832007
+14 163 0.0050103153551429415
+14 196 7.368110816386678e-05
+14 197 0.0014736221632773356
+14 198 0.0030209254347185383
+14 199 0.0009578544061302684
+14 206 7.368110816386678e-05
+14 246 0.0013262599469496023
+14 247 0.0061155319776009425
+14 269 7.368110816386678e-05
+14 276 0.0034630120837017397
+14 277 0.0008841732979664015
+14 290 0.001399941055113469
+14 291 0.0052313586796345415
+14 292 0.0058944886531093425
+14 293 0.008989095195991748
+14 468 0.0199675803124079
+14 469 0.02460949012673151
+14 470 0.021220159151193633
+14 471 0.02586206896551724
+14 472 0.020704391394046565
+14 473 0.017978190391983492
+14 474 0.020114942528735632
+14 475 0.02586206896551724
+14 476 0.02291482463896257
+14 477 0.02475685234305924
+14 478 0.021293840259357502
+14 479 0.026009431181844976
+14 480 0.019451812555260833
+14 481 0.014294134983790155
+14 482 0.01422045387562629
+14 483 0.02726201002063071
+14 484 0.02026230474506337
+14 485 0.015694076038903628
+14 486 0.02726201002063071
+14 487 0.02733569112879458
+14 488 0.01215738284703802
+14 489 0.009652225169466549
+14 490 0.015767757147067494
+14 491 0.02460949012673151
+14 492 0.020114942528735635
+14 493 0.013704686118479222
+14 494 0.01333628057765989
+14 495 0.022988505747126436
+14 496 0.018272914824638966
+14 497 0.020851753610374304
+14 498 0.016578249336870028
+14 499 0.025567344532861774
+14 500 0.007515473032714411
+14 501 0.019157088122605366
+14 502 0.015104627173592693
+14 503 0.00987326849395815
+14 504 0.021293840259357502
+14 505 0.020999115826702035
+14 506 0.013262599469496024
+14 507 0.013483642793987621
+14 508 0.010389036251105217
+14 509 0.011715296198054817
+14 510 0.010167992926613616
+14 511 0.011199528440907752
+14 512 0.009357500736811082
+14 513 0.020335985853227233
+14 514 0.010683760683760684
+14 515 0.01215738284703802
+14 516 0.016357206012378427
+14 517 0.004052460949012673
+14 518 0.006704980842911877
+14 519 0.004273504273504274
+14 520 0.0036103743000294726
+14 521 0.004494547597995874
+14 522 0.003020925434718538
+14 523 0.002136752136752137
+14 524 0.0037577365163572064
+14 525 0.0005894488653109342
+14 526 0.0008104921898025347
+14 531 0.0002947244326554671
+14 541 0.0016209843796050694
+14 542 0.0006631299734748011
+14 551 0.0019157088122605363
+14 552 0.0009578544061302684
+14 563 0.005010315355142941
+14 564 0.004715590922487474
+14 565 0.0010315355142941351
+14 567 0.000663129973474801
+14 568 0.00022104332449160037
+14 579 0.006115531977600943
+15 468 0.01103996467211305
+15 469 0.010230367262824759
+15 470 0.0023551924633841174
+15 471 0.004121586810922205
+15 472 0.009199970560094207
+15 473 0.011334363730036065
+15 474 0.004047987046441452
+15 475 0.0027967910502686394
+15 476 0.0059615809229410476
+15 477 0.0014719952896150733
+15 478 0.0003679988224037683
+15 479 0.0016191948185765807
+15 482 0.011187164201074557
+15 483 0.0056671818650180315
+15 484 0.0014719952896150733
+15 485 0.0003679988224037683
+15 486 0.004563185397806727
+15 487 0.0073599764480753675
+15 491 0.002134393169941856
+15 492 0.0003679988224037683
+15 495 0.0011775962316920587
+15 496 0.0005151983513652757
+15 497 0.0005151983513652757
+15 498 7.359976448075367e-05
+15 499 0.0032383896371531613
+15 500 0.019945536174284243
+15 501 0.01781114300434239
+15 502 0.014204754544785456
+15 503 0.02524471921689851
+15 504 0.005446382571575771
+15 505 0.010524766320747773
+15 506 0.01832634135570766
+15 507 0.01884153970707294
+15 508 0.018473540884669168
+15 511 0.01781114300434239
+15 512 0.019356738058438214
+15 513 0.012143961139324354
+15 514 0.020755133583572533
+15 515 0.01862074041363068
+15 516 0.015014351954073748
+15 517 0.024361522043129462
+15 518 0.02333112534039891
+15 519 0.027011113564436594
+15 520 0.02465592110105248
+15 521 0.024067122985206444
+15 522 0.024508721572090966
+15 523 0.023478324869360415
+15 524 0.025980716861706044
+15 525 0.018031942297784646
+15 526 0.020607934054611025
+15 527 0.0012511959961728123
+15 528 0.0059615809229410476
+15 529 0.0025023919923456246
+15 530 0.0009567969382497977
+15 531 0.018179141826746157
+15 532 0.006255979980864061
+15 533 0.011187164201074557
+15 534 0.005225583278133511
+15 535 0.004710384926768235
+15 536 0.0016927945830573343
+15 537 0.0007359976448075366
+15 538 0.013247957606535658
+15 540 7.359976448075367e-05
+15 541 0.02340472510487966
+15 542 0.02031353499668801
+15 543 0.010745565614190034
+15 544 0.0032383896371531613
+15 545 0.0003679988224037683
+15 546 0.015529550305439023
+15 547 0.005593582100537278
+15 548 0.001103996467211305
+15 549 0.019356738058438214
+15 550 0.009126370795613454
+15 551 0.025465518510340766
+15 552 0.022374328402149115
+15 553 0.0029439905792301465
+15 557 0.011923161845882095
+15 558 0.0029439905792301465
+15 559 0.00942076985353647
+15 560 0.003679988224037683
+15 561 0.0002943990579230147
+15 563 0.019356738058438214
+15 564 0.024582321336571723
+15 565 0.02244792816662987
+15 566 0.015382350776477514
+15 567 0.019503937587399718
+15 568 0.015161551483035255
+15 569 0.0059615809229410476
+15 570 0.01023036726282476
+15 571 0.0030175903437109006
+15 572 0.003459188930595423
+15 573 0.005519982336056524
+15 574 0.0008095974092882903
+15 575 0.0008095974092882903
+15 576 7.359976448075367e-05
+15 577 0.0013247957606535659
+15 578 0.0008095974092882903
+16 556 1.0
+17 17 0.0004919184820801125
+17 18 0.0006324666198172875
+17 20 0.005762473647224175
+17 21 0.0021082220660576245
+17 22 0.0014757554462403375
+17 23 0.0024595924104005625
+17 70 0.000140548137737175
+17 71 0.000983836964160225
+17 72 0.0023190442726633872
+17 73 0.0004919184820801125
+17 76 0.000140548137737175
+17 77 0.0006324666198172875
+17 80 0.008151791988756148
+17 81 0.006676036542515813
+17 82 0.016303583977512297
+17 83 0.012297962052002813
+17 96 0.002178496134926213
+17 97 0.0007027406886858749
+17 98 7.02740688685875e-05
+17 99 0.0011243851018974
+17 100 0.009065354884047786
+17 101 0.007308503162333099
+17 102 0.015038650737877725
+17 103 0.017919887561489812
+17 141 0.0018271257905832748
+17 142 0.003794799718903725
+17 148 0.0013352073085031624
+17 153 0.001546029515108925
+17 154 0.0024595924104005625
+17 160 0.0134926212227688
+17 161 0.01883345045678145
+17 162 0.012438510189739986
+17 163 0.005200281096275476
+17 178 0.0007730147575544624
+17 179 0.0007730147575544624
+17 180 0.003021784961349263
+17 181 0.00758959943780745
+17 182 0.01377371749824315
+17 183 0.007238229093464512
+17 184 0.0026001405481377374
+17 196 0.0123682361208714
+17 197 0.007449051300070275
+17 198 0.0071679550245959235
+17 199 0.0202389318341532
+17 200 0.0004919184820801125
+17 201 0.027406886858749122
+17 202 0.020028109627547436
+17 206 0.0019676739283204497
+17 207 0.0004919184820801125
+17 218 0.000140548137737175
+17 220 0.00028109627547435
+17 247 7.02740688685875e-05
+17 256 0.000140548137737175
+17 257 0.0004919184820801125
+17 269 0.00035137034434293746
+17 275 0.005270555165144062
+17 276 0.0010541110330288123
+17 277 7.02740688685875e-05
+17 278 0.02508784258608574
+17 289 0.019465917076598734
+17 290 0.0044975404075896
+17 291 0.001546029515108925
+17 292 0.0002108222066057625
+17 293 7.02740688685875e-05
+17 484 0.0009135628952916374
+17 485 0.0007730147575544624
+17 489 0.000421644413211525
+17 491 0.0004919184820801125
+17 492 7.02740688685875e-05
+17 493 0.0013352073085031622
+17 494 0.002951510892480675
+17 509 0.0002108222066057625
+17 510 0.0033731553056922
+17 579 0.0027406886858749122
+17 580 0.003162333099086437
+17 581 0.0023190442726633877
+17 582 0.009978917779339425
+17 583 0.009065354884047788
+17 584 0.0010541110330288125
+17 585 0.0016865776528460997
+17 586 0.004356992269852425
+17 587 0.003513703443429375
+17 588 0.0016865776528461
+17 589 0.004567814476458187
+17 590 0.009065354884047786
+17 591 0.006886858749121575
+17 592 0.013914265635980324
+17 593 0.016795502459592413
+17 594 0.021503865073787775
+17 595 0.028742094167252288
+17 596 0.02178496134926212
+17 597 0.02059030217849614
+17 598 0.0026001405481377374
+17 599 0.004427266338721012
+17 600 0.015038650737877721
+17 601 0.015390021082220663
+17 602 0.008081517919887564
+17 603 0.007308503162333099
+17 604 0.021995783555867888
+17 605 0.02312016865776529
+17 606 0.019465917076598734
+17 607 0.025720309205903027
+17 608 0.019465917076598737
+17 609 0.0123682361208714
+17 610 0.013070976809557275
+17 611 0.02009838369641603
+17 612 0.019184820801124384
+17 613 0.012719606465214337
+17 614 0.0134926212227688
+17 615 0.0179901616303584
+17 616 0.015319747013352071
+17 617 0.005621925509486999
+17 618 0.00028109627547435
+17 619 0.0022487702037948
+17 620 0.0026704146170063252
+17 621 0.0002108222066057625
+17 622 0.007589599437807451
+17 623 0.0018271257905832748
+17 624 0.00084328882642305
+17 625 0.0007027406886858749
+17 626 0.0009135628952916376
+17 627 0.01981728742094167
+17 628 0.0002108222066057625
+17 629 0.0004919184820801124
+17 630 0.003021784961349262
+17 631 0.0006324666198172875
+17 632 7.02740688685875e-05
+17 633 7.02740688685875e-05
+17 696 0.021152494729444835
+17 769 0.0179901616303584
+17 770 0.020941672522839076
+17 771 0.011243851018974
+17 772 0.005411103302881238
+17 773 0.0028109627547434997
+17 774 0.0036542515811665496
+17 775 0.00871398453970485
+17 776 0.012157413914265636
+17 777 0.00035137034434293746
+18 82 0.0006012777151446825
+18 83 0.00015031942878617063
+18 103 0.0004509582863585119
+18 142 7.515971439308531e-05
+18 160 0.001428034573468621
+18 161 0.0071401728673431055
+18 162 0.0018789928598271326
+18 182 0.0005261180007515971
+18 196 0.0018789928598271326
+18 197 0.0009019165727170237
+18 198 0.0008267568583239384
+18 199 0.006238256294626081
+18 201 0.003006388575723412
+18 202 0.0018038331454340473
+18 275 0.00015031942878617063
+18 278 0.002179631717399474
+18 289 0.003607666290868095
+18 580 0.020142803457346866
+18 581 0.01698609545283728
+18 582 0.03013904547162721
+18 583 0.02705749718151071
+18 584 0.017737692596768138
+18 585 0.016910935738444193
+18 586 0.021270199173243146
+18 587 0.018489289740698987
+18 588 0.01924088688462984
+18 589 0.016910935738444197
+18 590 0.016234498308906428
+18 591 0.017061255167230362
+18 592 0.015182262307403233
+18 593 0.010672679443818113
+18 594 0.004509582863585118
+18 595 0.005712138293874483
+18 596 0.01570838030815483
+18 597 0.011424276587748966
+18 598 0.018714768883878245
+18 599 0.01969184517098835
+18 600 0.020593761743705377
+18 601 0.023900789177001128
+18 602 0.027583615182262308
+18 603 0.0266065388951522
+18 604 0.007591131153701616
+18 605 0.007666290868094702
+18 606 0.012701991732431419
+18 607 0.020668921458098462
+18 608 0.016309658023299513
+18 609 0.015031942878617064
+18 610 0.017061255167230362
+18 611 0.009845922585494176
+18 612 0.009845922585494176
+18 613 0.02435174746335964
+18 614 0.02450206689214581
+18 615 0.02720781661029688
+18 616 0.016459977452085682
+18 617 0.021119879744456973
+18 618 0.012251033446072902
+18 619 0.020744081172491546
+18 620 0.02247275460353251
+18 621 0.008643367155204812
+18 622 0.025779782036828264
+18 623 0.02006764374295378
+18 624 0.011724915445321307
+18 625 0.01089815858699737
+18 626 0.011800075159714395
+18 627 0.006914693724163849
+18 628 0.011499436302142051
+18 629 0.011273957158962795
+18 630 0.0214956783164224
+18 631 0.012251033446072907
+18 632 0.013077790304396842
+18 633 0.014656144306651634
+18 634 0.00496054114994363
+18 635 0.006914693724163849
+18 636 0.004735062006764375
+18 637 0.004359263434798948
+18 638 0.004208944006012777
+18 639 0.003757985719654265
+18 640 0.0021796317173994736
+18 641 0.003908305148440436
+18 642 0.0010522360015031943
+18 643 0.0006012777151446825
+18 648 0.0005261180007515971
+18 658 0.0020293122886133035
+18 659 0.0011273957158962795
+18 668 0.000751597143930853
+18 669 0.0011273957158962795
+18 680 0.002931228861330327
+18 681 0.0057872980082675695
+18 682 0.001428034573468621
+18 684 0.0006012777151446825
+18 685 0.0006012777151446825
+18 696 0.003757985719654265
+18 769 0.0009019165727170238
+18 770 0.002931228861330327
+18 771 0.00015031942878617063
+18 775 0.00022547914317925594
+18 776 0.001202555430289365
+19 580 0.012027744982290436
+19 581 0.009961629279811098
+19 582 0.0059031877213695395
+19 583 0.004574970484061393
+19 584 0.009223730814639905
+19 585 0.011289846517119244
+19 586 0.0038370720188902006
+19 587 0.0028040141676505316
+19 588 0.0057556080283353
+19 589 0.0014757969303423849
+19 590 0.0003689492325855962
+19 591 0.0008116883116883117
+19 598 0.011806375442739079
+19 599 0.004501180637544274
+19 600 0.0012544273907910272
+19 601 0.0012544273907910272
+19 602 0.005165289256198347
+19 603 0.005165289256198347
+19 609 0.0005165289256198347
+19 610 0.0008116883116883117
+19 613 0.0016971664698937428
+19 614 0.0010330578512396697
+19 615 0.0011068476977567888
+19 616 0.0003689492325855962
+19 617 0.0028040141676505316
+19 618 0.021325265643447465
+19 619 0.012322904368358915
+19 620 0.016528925619834708
+19 621 0.022432113341204252
+19 622 0.0042060212514757975
+19 623 0.009518890200708384
+19 624 0.021989374262101534
+19 625 0.015053128689492327
+19 626 0.01977567886658796
+19 628 0.017635773317591502
+19 629 0.019406729634002362
+19 630 0.010256788665879575
+19 631 0.01977567886658796
+19 632 0.018299881936245575
+19 633 0.014684179456906728
+19 634 0.027007083825265645
+19 635 0.02317001180637544
+19 636 0.026638134592680048
+19 637 0.02368654073199528
+19 638 0.02435064935064935
+19 639 0.02457201889020071
+19 640 0.021915584415584416
+19 641 0.0256788665879575
+19 642 0.018816410861865408
+19 643 0.01682408500590319
+19 644 0.002656434474616293
+19 645 0.009445100354191263
+19 646 0.0031729634002361272
+19 647 0.0025826446280991736
+19 648 0.016602715466351833
+19 649 0.006419716646989375
+19 650 0.010478158205430934
+19 651 0.004870129870129871
+19 652 0.003246753246753247
+19 653 0.0014757969303423849
+19 654 0.0013282172373081465
+19 655 0.010847107438016527
+19 656 0.0005903187721369539
+19 657 0.0005165289256198347
+19 658 0.02169421487603306
+19 659 0.019406729634002362
+19 660 0.007747933884297522
+19 661 0.001844746162927981
+19 662 0.00014757969303423848
+19 663 0.012101534828807558
+19 664 0.0038370720188902014
+19 665 0.0007378984651711924
+19 666 0.0157172373081464
+19 667 0.006050767414403779
+19 668 0.02221074380165289
+19 669 0.021103896103896107
+19 670 0.0028040141676505316
+19 674 0.006419716646989373
+19 675 0.0014020070838252656
+19 676 0.010109208972845335
+19 677 0.0030253837072018895
+19 678 7.378984651711924e-05
+19 679 0.00014757969303423848
+19 680 0.02206316410861866
+19 681 0.022432113341204252
+19 682 0.0256788665879575
+19 683 0.01977567886658796
+19 684 0.022358323494687134
+19 685 0.02088252656434475
+19 686 0.012470484061393153
+19 687 0.01586481700118064
+19 688 0.004427390791027155
+19 689 0.006198347107438017
+19 690 0.00974025974025974
+19 691 0.000885478158205431
+19 692 0.0003689492325855962
+19 693 0.00014757969303423848
+19 694 0.002877804014167651
+19 695 0.0016233766233766235
+20 673 1.0

data/body_models/J_regressor_mano_RIGHT.txt

@@ -0,0 +1,1902 @@
+# 21 778
+0 4 0.0019103600293901542
+0 5 0.0027920646583394562
+0 6 0.00029390154298310065
+0 7 0.00014695077149155033
+0 25 0.0016164584864070536
+0 26 0.000440852314474651
+0 32 0.011756061719324026
+0 33 0.021234386480529024
+0 34 0.019838354151359296
+0 35 0.016311535635562088
+0 36 0.015870683321087434
+0 37 0.02343864805290228
+0 38 0.01671565025716385
+0 39 0.020499632623071272
+0 40 0.005437178545187362
+0 41 0.010139603232916973
+0 42 0.002645113886847906
+0 43 0.00014695077149155033
+0 44 0.02005878030859662
+0 45 0.02233651726671565
+0 50 0.01763409257898604
+0 51 0.01704628949301984
+0 52 0.019838354151359296
+0 53 0.02079353416605437
+0 54 0.00822924320352682
+0 55 0.00822924320352682
+0 78 0.011572373254959589
+0 79 0.011939750183688464
+0 84 0.01704628949301984
+0 85 0.019691403379867745
+0 88 0.005437178545187362
+0 89 0.0007347538574577516
+0 90 0.014548126377663484
+0 91 0.018736223365172666
+0 92 0.011645848640705364
+0 106 0.018515797207935343
+0 107 0.02204261572373255
+0 108 0.012417340191036004
+0 109 0.009992652461425423
+0 110 0.016311535635562088
+0 111 0.01880969875091844
+0 112 0.0073475385745775165
+0 113 0.0014695077149155032
+0 114 0.005731080088170463
+0 116 0.02204261572373255
+0 117 0.012123438648052902
+0 118 0.013005143277002204
+0 119 0.016385011021307863
+0 120 0.008155767817781044
+0 121 0.011315209404849376
+0 122 0.009037472446730345
+0 130 0.0073475385745775165
+0 131 0.00911094783247612
+0 178 0.001763409257898604
+0 179 0.002351212343864805
+0 190 0.019544452608376194
+0 191 0.019691403379867745
+0 192 0.01704628949301984
+0 193 0.016605437178545186
+0 200 0.002351212343864805
+0 203 0.00822924320352682
+0 204 0.007641440117560617
+0 205 0.01704628949301984
+0 207 0.001763409257898604
+0 208 0.005290227773695812
+0 209 0.01763409257898604
+0 210 0.019691403379867745
+0 211 0.019691403379867745
+0 214 0.011315209404849376
+0 215 0.011315209404849376
+0 216 0.007641440117560617
+0 217 0.00822924320352682
+0 218 0.002351212343864805
+0 219 0.0011756061719324026
+0 227 0.002351212343864805
+0 229 0.007788390889052168
+0 231 0.002204261572373255
+0 232 0.016311535635562088
+0 233 0.006759735488611315
+0 234 0.011168258633357825
+0 235 0.019544452608376194
+0 236 0.0016164584864070536
+0 239 0.011315209404849376
+0 241 0.0007347538574577516
+0 242 0.002351212343864805
+0 243 0.0036737692872887582
+0 244 0.0011756061719324026
+0 254 0.0064658339456282144
+0 255 0.0038207200587803084
+0 256 0.002351212343864805
+0 257 0.002351212343864805
+0 264 0.014107274063188832
+0 265 0.00440852314474651
+0 279 0.011315209404849376
+0 284 0.00896399706098457
+0 285 0.0029390154298310064
+1 0 0.014595751184471957
+1 1 0.025294207550053488
+1 2 0.019180803912578332
+1 3 0.01039278618370778
+1 4 0.03156044627846554
+1 5 0.025752712822864135
+1 6 0.014977838911814154
+1 7 0.023307351367874065
+1 8 0.005654898364664528
+1 9 0.009170105456212748
+1 10 0.002063273727647868
+1 11 0.0006113403637475165
+1 12 0.0018340210912425497
+1 14 0.001222680727495033
+1 15 7.641754546843957e-05
+1 16 0.0011462631820265935
+1 17 0.0004585052728106374
+1 18 0.00015283509093687913
+1 19 0.0003820877273421978
+1 22 7.641754546843957e-05
+1 24 0.01413724591166132
+1 25 0.019257221458046772
+1 26 0.024377197004432218
+1 27 0.017346782821335782
+1 28 0.0007641754546843956
+1 29 0.0022161088185847473
+1 30 0.0006877579092159561
+1 31 0.0005349228182790769
+1 32 0.0005349228182790768
+1 33 0.0005349228182790769
+1 34 0.0024071526822558465
+1 35 0.002445361454990066
+1 36 0.029802842732691428
+1 37 0.022122879413113253
+1 38 0.010029802842732692
+1 39 0.02334556014060829
+1 40 0.029344337459880795
+1 41 0.032171786642213054
+1 42 0.02009781445819961
+1 43 0.009934280910897143
+1 60 0.004355800091701055
+1 61 0.00855876509246523
+1 62 0.0004585052728106374
+1 63 0.003285954455142901
+1 64 0.0012990982729634726
+1 65 7.641754546843957e-05
+1 66 0.0019868561821794286
+1 67 0.004814305364511693
+1 68 0.008253094910591475
+1 69 0.0018340210912425497
+1 70 0.0003820877273421978
+1 71 7.641754546843957e-05
+1 88 0.021320495185694635
+1 89 0.013907993275256002
+1 90 0.01986856182179429
+1 91 0.013564114320648022
+1 92 0.003763564114320649
+1 93 0.0004585052728106374
+1 94 0.008329512456059913
+1 95 0.007565337001375517
+1 104 0.0027510316368638244
+1 105 0.0072596668195017595
+1 109 0.009705028274491823
+1 110 0.005654898364664528
+1 111 0.015436344184624792
+1 112 0.019180803912578332
+1 113 0.03339446736970809
+1 114 0.0340058077334556
+1 115 0.02559987773192725
+1 116 0.008405930001528351
+1 117 0.0017767079321412199
+1 118 0.00527281063732233
+1 119 0.00032477456824086816
+1 122 0.004967140455448571
+1 123 0.007259666819501758
+1 124 0.0016811860003056705
+1 125 0.0025217790004585057
+1 126 0.008176677365123033
+1 129 0.00030567018187375826
+1 145 0.00030567018187375826
+1 146 0.0006877579092159561
+1 147 7.641754546843957e-05
+1 152 7.641754546843957e-05
+1 157 0.002063273727647868
+1 158 0.0016047684548372307
+1 159 0.0032095369096744614
+1 188 0.0007641754546843956
+1 190 0.0019868561821794286
+1 191 0.0004585052728106374
+1 192 0.0016047684548372307
+1 193 0.005884151001069847
+1 207 0.00015283509093687913
+1 208 7.641754546843957e-05
+1 209 0.00030567018187375826
+1 216 0.0008405930001528353
+1 217 0.003897294818890417
+1 218 0.0008405930001528353
+1 219 0.0014519333639003516
+1 227 0.005502063273727648
+1 229 0.008635182637933671
+1 230 0.004126547455295736
+1 231 0.009705028274491824
+1 232 0.01245605991135565
+1 233 0.016888277548525142
+1 234 0.001413724591166132
+1 235 0.005654898364664528
+1 236 0.012838147638697846
+1 239 0.00026746140913953847
+1 240 0.01543634418462479
+1 241 0.0006877579092159561
+1 242 0.0032095369096744614
+1 248 0.004890722909980132
+1 249 0.0005349228182790769
+1 250 0.0015283509093687911
+1 251 0.0009170105456212748
+1 252 0.0029038667278007036
+1 253 0.005502063273727649
+1 254 0.0019868561821794286
+1 255 0.0002292526364053187
+1 264 0.028885832187070158
+1 265 0.029650007641754548
+1 266 0.006953996637628001
+1 267 0.002445361454990066
+1 268 0.00015283509093687913
+1 285 0.010087116001834023
+1 286 0.007794589637780836
+1 287 0.0025981965459269452
+1 697 0.0004585052728106374
+1 699 7.641754546843957e-05
+1 700 0.00030567018187375826
+1 704 0.0002292526364053187
+1 705 0.0008405930001528353
+1 706 7.641754546843957e-05
+2 0 0.0027531810402559712
+2 1 0.0034972840241089364
+2 2 0.007887491628841432
+2 3 0.0056551826772825355
+2 4 0.009152466701391472
+2 5 0.01674231713669172
+2 6 0.02708534861224793
+2 7 0.02209985862043307
+2 8 0.00833395341915321
+2 9 0.009152466701391472
+2 10 0.011682416846491553
+2 11 0.0055063620805119425
+2 12 0.005431951782126646
+2 13 0.0011161544757794478
+2 14 0.006176054765979612
+2 15 0.0017858471612471167
+2 16 0.0007441029838529652
+2 19 0.0003720514919264826
+2 26 0.000967333879008855
+2 27 0.0008929235806235583
+2 28 0.013245033112582783
+2 29 0.013765905201279856
+2 30 0.009970979983629735
+2 31 0.011384775652950369
+2 36 0.0023811295483294886
+2 37 0.00014882059677059304
+2 38 7.441029838529652e-05
+2 39 0.0020834883547883026
+2 40 0.0055063620805119425
+2 41 0.009896569685244438
+2 42 0.022843961604286034
+2 43 0.032666120991145166
+2 60 0.00364610462087953
+2 61 0.0017858471612471167
+2 62 0.0002976411935411861
+2 63 0.000967333879008855
+2 64 0.0014882059677059304
+2 65 0.0004464617903117792
+2 68 0.0002976411935411861
+2 69 7.441029838529652e-05
+2 88 0.01562616266091227
+2 89 0.027234169209018527
+2 90 0.00513431058858546
+2 91 0.0006696926854676687
+2 93 7.441029838529652e-05
+2 94 0.0005952823870823722
+2 104 0.025225091152615526
+2 105 0.017858471612471165
+2 113 0.0035716943224942334
+2 114 0.002604360443485378
+2 115 0.010566262370712107
+2 123 0.026787707418706754
+2 124 0.021504576233350697
+2 125 0.01882580549148002
+2 126 0.02083488354788303
+2 127 0.0002232308951558896
+2 128 0.0002976411935411861
+2 129 0.0017114368628618197
+2 144 0.0002232308951558896
+2 145 0.0013393853709353374
+2 158 0.002604360443485378
+2 193 0.0003720514919264826
+2 217 0.0007441029838529652
+2 219 0.0004464617903117792
+2 227 0.003199642830567751
+2 229 0.003125232532182454
+2 230 0.008854825507850286
+2 231 0.00982215938685914
+2 232 0.002009078056403006
+2 233 0.007813081330456134
+2 235 7.441029838529652e-05
+2 236 0.01912344668502121
+2 240 0.01480764937867401
+2 248 0.03318699307984225
+2 249 0.01823052310439765
+2 250 0.02887119577349505
+2 251 0.02500186025745963
+2 252 0.02864796487833916
+2 253 0.032889351886301064
+2 259 0.00014882059677059304
+2 264 0.0002232308951558896
+2 265 0.0005952823870823722
+2 266 0.015402931765756382
+2 267 0.01622144504799464
+2 286 0.02805268249125679
+2 287 0.025820373539697895
+2 697 0.014510008185132822
+2 698 0.008631594612694398
+2 699 0.011161544757794479
+2 700 0.01049185207232681
+2 701 0.00811072252399732
+2 702 0.013393853709353377
+2 703 0.010938313862638589
+2 704 0.008185132822382618
+2 705 0.02187662772527718
+2 706 0.018825805491480024
+2 707 0.011905647741647447
+2 708 0.007217798943373763
+2 709 0.005059900290200163
+2 710 0.003199642830567751
+2 711 0.0019346677580177095
+2 712 0.005952823870823722
+2 713 0.00364610462087953
+2 714 0.00364610462087953
+2 715 0.0026787707418706747
+2 716 0.0021578986531735995
+2 721 0.0006696926854676687
+2 722 0.0002232308951558896
+2 723 0.0002232308951558896
+2 725 0.0004464617903117792
+2 731 0.0032740531289530473
+2 732 0.0008185132822382618
+2 741 0.0005952823870823722
+2 742 0.0005208720886970756
+2 746 0.0002232308951558896
+2 749 0.0005208720886970756
+2 753 0.0034972840241089364
+2 754 0.004018156112806012
+2 755 0.0014882059677059304
+2 757 0.0008929235806235583
+2 758 0.0014137956693206339
+2 759 0.0003720514919264826
+2 760 7.441029838529652e-05
+3 6 0.0019164148301024542
+3 7 0.0014004569912287167
+3 8 0.000884499152354979
+3 9 0.00029483305078499295
+3 10 0.004422495761774894
+3 11 0.0011793322031399718
+3 12 0.0005896661015699859
+3 14 0.0011056239404437236
+3 28 0.011203655929829732
+3 29 0.0037591213975086604
+3 30 0.004496204024471142
+3 31 0.011645905506007222
+3 43 0.0019164148301024544
+3 89 0.0005896661015699859
+3 104 0.009729490675904768
+3 105 0.002137539618191199
+3 123 0.006412618854573597
+3 124 0.0187956069875433
+3 125 0.013414903810717178
+3 126 0.004938453600648632
+3 230 0.0007370826269624824
+3 231 0.00022112478808874474
+3 236 0.0005159578388737376
+3 240 0.0008844991523549787
+3 248 0.007665659320409817
+3 249 0.013120070759932186
+3 250 0.009434657625119773
+3 251 0.012088155082184712
+3 252 0.004348787499078646
+3 253 0.003022038770546178
+3 266 0.0029483305078499295
+3 267 0.0125304046583622
+3 286 0.002727205719761185
+3 287 0.005896661015699859
+3 697 0.01805852436058082
+3 698 0.019016731775632047
+3 699 0.021375396181911987
+3 700 0.01968010613989828
+3 701 0.023512935800103187
+3 702 0.01975381440259453
+3 703 0.021965062283481978
+3 704 0.019164148301024544
+3 705 0.015331318640819633
+3 706 0.017837399572492075
+3 707 0.02889363897692931
+3 708 0.02130168791921574
+3 709 0.027050932409523103
+3 710 0.024544851477850665
+3 711 0.0209331466057345
+3 712 0.0232181027493182
+3 713 0.023070686223925697
+3 714 0.024102601901673175
+3 715 0.018353357411365814
+3 716 0.017026608682833344
+3 717 0.0016952900420137097
+3 718 0.0062652023291811
+3 719 0.0033168718213311705
+3 720 0.00125304046583622
+3 721 0.016879192157440846
+3 722 0.01090882287904474
+3 723 0.008402741947372299
+3 724 0.004717328812559887
+3 725 0.010982531141740989
+3 726 0.0033168718213311705
+3 727 0.0008107908896587306
+3 730 7.370826269624824e-05
+3 731 0.022775853173140702
+3 732 0.018279649148669565
+3 733 0.009803198938601014
+3 734 0.003022038770546178
+3 735 0.0003685413134812412
+3 736 0.011719613768703471
+3 737 0.003906537922901157
+3 738 0.0008107908896587306
+3 739 0.013488612073413427
+3 740 0.005306994914129874
+3 741 0.021301687919215745
+3 742 0.019606397877202027
+3 743 0.0022112478808874476
+3 746 0.006338910591877348
+3 747 0.00125304046583622
+3 748 0.0016952900420137097
+3 749 0.009876907201297264
+3 750 0.003022038770546178
+3 751 7.370826269624824e-05
+3 753 0.025208225842116898
+3 754 0.0209331466057345
+3 755 0.023291811012014444
+3 756 0.017837399572492075
+3 757 0.021449104444608236
+3 758 0.01975381440259453
+3 759 0.01171961376870347
+3 760 0.01348861207341343
+3 761 0.003906537922901157
+3 762 0.005306994914129872
+3 763 0.007960492371194809
+3 764 0.0008107908896587306
+3 765 0.0003685413134812412
+3 767 0.0022112478808874476
+3 768 0.0011056239404437238
+4 745 1.0
+5 0 0.0012638674343491084
+5 1 0.0001404297149276787
+5 2 0.00035107428731919675
+5 3 0.002808594298553574
+5 8 0.004072461732902682
+5 9 0.0007723634321022329
+5 10 0.004774610307541076
+5 11 0.01418340120769555
+5 12 0.012357814913635726
+5 13 0.01930908580255582
+5 14 0.007934278893413846
+5 15 0.020011234377194213
+5 16 0.0021064457239151806
+5 17 0.0006319337171745541
+5 18 0.0022468754388428594
+5 19 0.009127931470299114
+5 21 0.00042128914478303613
+5 24 0.0009127931470299115
+5 25 7.021485746383936e-05
+5 26 0.0001404297149276787
+5 27 0.0010532228619575903
+5 28 0.0004212891447830361
+5 29 0.0015447268642044658
+5 30 0.003932032017975004
+5 31 0.0009127931470299115
+5 46 0.0006319337171745542
+5 47 0.00035107428731919675
+5 48 0.003721387445583485
+5 49 0.0027383794410897346
+5 56 0.0002808594298553574
+5 57 7.021485746383936e-05
+5 58 0.0010532228619575903
+5 59 0.0028788091560174134
+5 60 0.010040724617329027
+5 61 0.005687403454570988
+5 62 0.029981744137059403
+5 63 0.017483499508496
+5 64 0.02029209380704957
+5 65 0.024294340682488414
+5 66 0.0029490240134812527
+5 67 0.0011234377194214297
+5 68 0.005827833169498665
+5 69 0.00975986518747367
+5 74 0.00217666058137902
+5 75 0.0010532228619575903
+5 76 0.00035107428731919675
+5 77 0.00021064457239151807
+5 86 0.0007723634321022329
+5 87 0.0021064457239151806
+5 93 0.018536722370453586
+5 94 0.0016851565791321445
+5 95 0.0001404297149276787
+5 104 7.021485746383936e-05
+5 105 0.0001404297149276787
+5 127 0.023592192107850022
+5 128 0.02710293498104199
+5 129 0.020713382951832608
+5 132 0.023030473248139307
+5 133 0.005195899452324112
+5 134 0.005195899452324112
+5 135 0.01305996348827412
+5 136 0.008495997753124563
+5 137 0.014323830922623225
+5 138 0.01818564808313439
+5 139 0.011515236624069652
+5 140 0.008215138323269205
+5 143 0.010742873191967421
+5 144 0.016991995506249125
+5 145 0.010040724617329027
+5 146 0.00035107428731919675
+5 147 0.0011234377194214297
+5 149 0.013832326920376354
+5 150 0.016430276646538407
+5 151 0.010181154332256704
+5 152 0.011023732621822779
+5 155 0.00035107428731919675
+5 156 0.001966016008987502
+5 157 7.021485746383936e-05
+5 158 0.003932032017975004
+5 164 0.0034405280157281284
+5 165 0.005195899452324111
+5 166 0.0014745120067406266
+5 167 0.0014745120067406264
+5 168 0.026049712119084405
+5 169 0.02927959556242101
+5 170 0.023873051537705376
+5 171 0.016008987501755372
+5 172 0.027102934981041993
+5 173 0.016921780648785283
+5 174 0.005546973739643309
+5 175 0.005406544024715631
+5 176 0.013551467490520995
+5 177 0.00758320460609465
+5 183 7.021485746383936e-05
+5 185 0.009127931470299114
+5 186 0.017834573795815194
+5 187 0.008074708608341525
+5 189 0.007161915461311614
+5 194 0.010602443477039742
+5 195 0.01060244347703974
+5 206 0.0013340822918129478
+5 212 0.007091700603847775
+5 213 0.0013340822918129476
+5 219 0.0002808594298553574
+5 220 0.00435332116275804
+5 222 0.0002808594298553574
+5 223 0.00042128914478303613
+5 225 0.0016851565791321445
+5 226 0.00042128914478303613
+5 227 0.000983008004493751
+5 228 0.00975986518747367
+5 230 0.001825586294059823
+5 231 7.021485746383936e-05
+5 246 0.00035107428731919675
+5 258 0.020924027524224127
+5 259 0.022398539530964757
+5 260 0.015587698356972338
+5 261 0.012568459486027245
+5 262 0.009619435472545991
+5 263 0.01305996348827412
+5 266 0.0010532228619575903
+5 267 0.0005617188597107148
+5 268 0.004283106305294201
+5 269 0.0017553714365959837
+5 270 0.005266114309787951
+5 271 0.004844825165004915
+5 274 0.018045218368206713
+5 276 0.0002808594298553574
+5 277 0.00021064457239151807
+5 280 0.0001404297149276787
+5 288 0.00540654402471563
+5 290 7.021485746383936e-05
+5 358 0.0002808594298553574
+5 359 0.00035107428731919675
+5 362 0.00021064457239151807
+5 363 0.0002808594298553574
+5 365 7.021485746383936e-05
+5 366 0.0009127931470299116
+5 367 0.0013340822918129476
+5 368 0.005125684594860273
+5 369 0.0034405280157281284
+5 370 0.0013340822918129476
+5 371 0.00021064457239151807
+5 373 0.00042128914478303613
+5 375 0.00035107428731919675
+5 378 0.004493750877685719
+5 379 0.0034405280157281284
+5 380 0.004634180592613397
+5 383 0.00042128914478303613
+5 385 0.0016149417216683051
+5 386 0.001404297149276787
+5 387 0.0016851565791321445
+5 388 0.0002808594298553574
+5 399 0.0014745120067406264
+6 46 0.019904998869034157
+6 47 0.01960340797707909
+6 48 0.025559828093191583
+6 49 0.02352408957249491
+6 56 0.022166930558697125
+6 57 0.020131192038000453
+6 58 0.02194073738973083
+6 59 0.028952725627686037
+6 62 0.0005277840609213601
+6 65 0.00022619316896629722
+6 86 0.02382568046444997
+6 87 0.022543919173640955
+6 127 0.0012063635678202518
+6 128 0.0007539772298876573
+6 132 0.0006031817839101259
+6 133 0.017643067179371183
+6 134 0.02382568046444997
+6 135 0.01379778330694413
+6 136 0.01259141973912388
+6 137 0.004448465656337178
+6 138 0.003091306642539395
+6 139 0.009424715373595717
+6 140 0.012214431124180048
+6 143 0.0005277840609213601
+6 144 0.0012817612908090175
+6 150 0.0008293749528764231
+6 155 0.019678805700067855
+6 156 0.0244288622483601
+6 164 0.019980396592022914
+6 165 0.017944658071326246
+6 166 0.023222498680539848
+6 167 0.023901078187438737
+6 168 0.002789715750584332
+6 169 0.002186533966674206
+6 170 0.00987710171152831
+6 171 0.005881022393123726
+6 172 0.004071477041393349
+6 173 0.011837442509236221
+6 174 0.022166930558697128
+6 175 0.02382568046444997
+6 176 0.019377214808112796
+6 177 0.013119203800045236
+6 185 0.0016587499057528462
+6 186 0.004448465656337178
+6 187 0.0005277840609213601
+6 189 0.020809771544899342
+6 194 0.015154942320741913
+6 195 0.01839704440925884
+6 212 0.021262157882831936
+6 213 0.022317726004674656
+6 221 0.006333408731056322
+6 222 0.016210510442584633
+6 223 0.018472442132247607
+6 224 0.00987710171152831
+6 225 0.02744477116791073
+6 226 0.020583578375933047
+6 228 0.0005277840609213602
+6 237 0.012516022016135112
+6 238 0.011912840232224985
+6 245 0.011912840232224985
+6 258 0.0052024428862248355
+6 259 0.002337329412651738
+6 260 0.007162783683932745
+6 261 0.013043806077056472
+6 262 0.0016587499057528462
+6 263 0.007388976852899043
+6 272 0.014174771921887958
+6 273 0.012817612908090177
+6 274 0.0059564201161124925
+6 280 0.019301817085124028
+6 281 0.011385056171303627
+6 282 0.011460453894292393
+6 283 0.017643067179371186
+6 294 0.003920681595415819
+6 295 0.0069365905149664465
+6 296 0.0037698861494382865
+6 297 0.00512704516323607
+6 298 0.006634999623011385
+6 299 0.002789715750584332
+6 300 0.0021865339666742064
+6 301 0.0038452838724270517
+6 302 0.0005277840609213601
+6 303 0.0006031817839101259
+6 305 0.00030159089195506294
+6 316 0.0016587499057528462
+6 321 0.0009047726758651889
+6 330 0.0021111362436854408
+6 331 0.0015079544597753145
+6 340 0.00512704516323607
+6 341 0.004599261102314709
+6 342 0.0011309658448314859
+6 344 0.0007539772298876573
+6 345 0.00022619316896629722
+7 46 0.008690077640857611
+7 47 0.009188688653037966
+7 48 0.0033478167960680964
+7 49 0.0034902770852624832
+7 56 0.010898212123370611
+7 57 0.012322815015314481
+7 58 0.004202578531234419
+7 59 0.003276586651470902
+7 86 0.00648194315834461
+7 87 0.0016382933257354513
+7 133 0.00035615072298596765
+7 134 0.0015670631811382577
+7 155 0.009829759954412709
+7 156 0.004131348386637225
+7 164 0.0009259918797635161
+7 165 0.0006410713013747418
+7 166 0.003917657952845645
+7 167 0.0050573402664007405
+7 174 0.001638293325735451
+7 175 0.0014246028919438706
+7 189 0.0009259918797635161
+7 194 0.00028492057838877413
+7 195 0.0006410713013747418
+7 212 0.00042738086758316123
+7 213 0.0037039675190540643
+7 221 0.019517059619631027
+7 222 0.016739083980340477
+7 223 0.0143172590640359
+7 224 0.02443193959683738
+7 225 0.00683809388133058
+7 226 0.01111190255716219
+7 237 0.016739083980340477
+7 238 0.018092456727687157
+7 245 0.01367618776266116
+7 272 0.02236626540351877
+7 273 0.01923213904124225
+7 280 0.011040672412564997
+7 281 0.020086900776408578
+7 282 0.01859106773986751
+7 283 0.0165253935465489
+7 294 0.024004558729254222
+7 295 0.024075788873851416
+7 296 0.02443193959683738
+7 297 0.025357931476600898
+7 298 0.026283923356364414
+7 299 0.023933328584657028
+7 300 0.022722416126504736
+7 301 0.02514424104280932
+7 302 0.01738015528171522
+7 303 0.020941662511574897
+7 304 0.007835315905691288
+7 305 0.017380155281715225
+7 306 0.011396823135550965
+7 307 0.0036327373744568705
+7 308 0.0012821426027494836
+7 309 0.002777975639290548
+7 310 0.011966664292328516
+7 311 0.005342260844789515
+7 312 0.0038464278082484507
+7 313 0.0014958330365410642
+7 314 0.0007835315905691288
+7 315 0.008191466628677256
+7 316 0.022651185981907542
+7 317 0.00035615072298596765
+7 321 0.02101289265617209
+7 322 0.01225158487071729
+7 323 0.007764085761094094
+7 324 0.002564285205498967
+7 325 0.01994444048721419
+7 326 0.008690077640857611
+7 327 0.0024218249163045803
+7 328 0.0165253935465489
+7 329 0.006980554170524965
+7 330 0.028064676971294254
+7 331 0.021084122800769284
+7 332 0.0019232139041242254
+7 333 0.00021369043379158061
+7 334 0.010969442267967804
+7 335 0.0024930550609017737
+7 336 0.008690077640857611
+7 337 0.003988888097442838
+7 338 0.00028492057838877413
+7 340 0.019588289764228224
+7 341 0.0242182491630458
+7 342 0.021867654391338417
+7 343 0.014103568630244322
+7 344 0.018662297884464708
+7 345 0.014673409787021868
+7 346 0.006125792435358643
+7 347 0.009758529809815513
+7 348 0.0017095234703326447
+7 349 0.0031341263622765153
+7 350 0.004772419688011967
+7 351 0.0006410713013747418
+7 352 0.0008547617351663223
+7 353 0.00042738086758316123
+7 354 0.001068452168957903
+7 355 0.0009972220243607095
+8 317 1.0
+9 11 0.0002498906728306366
+9 13 0.0002498906728306366
+9 14 0.0009995626913225464
+9 15 0.0022490160554757294
+9 16 0.0029986880739676387
+9 17 0.002249016055475729
+9 18 0.007746610857749733
+9 19 0.00949584556756419
+9 20 0.0013743987005685012
+9 21 0.00437308677453614
+9 22 0.0009995626913225461
+9 23 0.00018741800462297744
+9 48 0.0004997813456612732
+9 59 0.0002498906728306366
+9 62 0.0014368713687761604
+9 63 0.000874617354907228
+9 64 6.247266820765915e-05
+9 65 6.247266820765915e-05
+9 66 0.0024989067283063657
+9 67 0.000437308677453614
+9 68 0.0006871993502842506
+9 69 0.0029986880739676387
+9 71 0.0004997813456612732
+9 74 0.015555694383707127
+9 75 0.017867183107390515
+9 76 0.017242456425313923
+9 77 0.00868370088086462
+9 83 6.247266820765915e-05
+9 87 0.0004997813456612732
+9 93 0.0033110514150059348
+9 127 0.0006247266820765914
+9 132 0.004810395451989753
+9 133 0.0006247266820765914
+9 135 0.0001249453364153183
+9 136 0.0004997813456612732
+9 137 0.015555694383707127
+9 138 0.007246829512088461
+9 139 0.005997376147935278
+9 140 0.008683700880864622
+9 141 0.005997376147935278
+9 142 0.0025613793965140247
+9 143 0.015743112388330104
+9 144 0.009558318235771848
+9 145 0.0032485787467982754
+9 146 0.0015618167051914785
+9 147 0.006122321484350596
+9 148 0.0025613793965140247
+9 149 0.0071843568438808006
+9 150 0.01243206097332417
+9 151 0.013993877678515648
+9 152 0.007809083525957393
+9 157 0.0001249453364153183
+9 158 0.0023114887236833884
+9 160 0.0019991253826450927
+9 161 0.0002498906728306366
+9 162 0.0005622540138689324
+9 163 0.0021240707190604106
+9 164 0.0029362154057599797
+9 165 0.002561379396514025
+9 166 0.0007496720184919098
+9 167 0.0007496720184919097
+9 168 0.002124070719060411
+9 169 0.0003123633410382957
+9 170 0.0006871993502842506
+9 171 0.002249016055475729
+9 174 0.0028737427375523207
+9 175 0.0018741800462297744
+9 176 0.009433372899356529
+9 177 0.006247266820765914
+9 181 0.00018741800462297744
+9 182 0.0009995626913225464
+9 183 0.004248141438120822
+9 185 0.019179109139751356
+9 186 0.01661772974323733
+9 187 0.019054163803336036
+9 194 0.0015618167051914785
+9 195 0.0001249453364153183
+9 196 0.0004997813456612732
+9 197 0.0014993440369838195
+9 198 0.0003748360092459549
+9 199 0.0001249453364153183
+9 202 6.247266820765915e-05
+9 206 0.013181732991816079
+9 207 6.247266820765915e-05
+9 212 0.0018741800462297742
+9 213 0.0002498906728306366
+9 218 0.0003123633410382957
+9 219 0.0006871993502842506
+9 220 0.014868495033422876
+9 225 0.0006247266820765914
+9 227 0.0006871993502842506
+9 228 0.021802961204473042
+9 230 0.0002498906728306366
+9 246 0.020803398513150495
+9 247 0.017304929093521583
+9 258 0.0004997813456612732
+9 259 0.0027487974011370024
+9 260 0.0017492347098144558
+9 261 0.002623852064721684
+9 262 0.01974136315362029
+9 263 0.01655525707502967
+9 268 0.007746610857749734
+9 269 0.02167801586805772
+9 270 0.019054163803336036
+9 271 0.011932279627662898
+9 274 0.0066221028300118695
+9 275 0.0007496720184919098
+9 276 0.016742675079652648
+9 277 0.02205285187730368
+9 288 0.022427687886549634
+9 289 0.0003123633410382957
+9 290 0.00730930218029612
+9 291 0.005685012806896982
+9 292 0.0057474854751046415
+9 293 0.008933591553695257
+9 356 0.0014993440369838195
+9 357 0.0014993440369838193
+9 358 0.00668457549821953
+9 359 0.004685450115574436
+9 360 0.0007496720184919098
+9 361 0.0007496720184919098
+9 362 0.0024989067283063657
+9 363 0.0038733054288748667
+9 364 0.0014368713687761604
+9 365 0.004498032110951459
+9 366 0.009933154245017804
+9 367 0.010245517586056099
+9 368 0.015993003061160742
+9 369 0.015993003061160742
+9 370 0.021115761854188793
+9 371 0.01693009308427563
+9 372 0.0009995626913225464
+9 373 0.0037483600924595483
+9 374 0.008996064221902918
+9 375 0.012432060973324168
+9 376 0.004498032110951458
+9 377 0.0031861060785906164
+9 378 0.017554819766352217
+9 379 0.01749234709814456
+9 380 0.01649278440682201
+9 381 0.008308864871618667
+9 382 0.006434684825388891
+9 383 0.016055475729368402
+9 384 0.012557006309739488
+9 385 0.01018304491784844
+9 386 0.015180858374461174
+9 387 0.01155744361841694
+9 388 0.009058536890110576
+9 389 0.0028112700693446614
+9 391 0.00018741800462297744
+9 392 0.0005622540138689324
+9 394 0.0018117073780221152
+9 395 0.0004997813456612732
+9 399 0.01611794839757606
+9 402 0.0008746173549072279
+9 470 0.0007496720184919098
+9 471 0.0004997813456612732
+9 478 0.0007496720184919098
+9 479 0.0004997813456612732
+9 480 0.0026863247329293434
+9 481 0.002623852064721684
+9 483 0.0001249453364153183
+9 484 0.0001249453364153183
+9 485 0.0014993440369838195
+9 486 0.0004997813456612732
+9 488 0.008996064221902916
+9 489 0.006059848816142937
+9 490 0.006497157493596552
+9 491 0.0001249453364153183
+9 492 0.0003748360092459549
+9 493 0.001311926032360842
+9 494 0.000437308677453614
+9 495 0.0017492347098144558
+9 496 0.002623852064721684
+9 497 0.0027487974011370024
+9 498 0.0006247266820765914
+9 509 0.0020615980508527517
+9 510 0.0003748360092459549
+9 579 0.0019991253826450927
+10 74 0.0005264345341054373
+10 75 0.0021809430698653833
+10 76 0.000752049334436339
+10 137 0.000827254267879973
+10 143 0.0006016394675490712
+10 150 0.0003008197337745356
+10 151 0.0006016394675490712
+10 185 0.004361886139730767
+10 186 0.0010528690682108748
+10 187 0.003910656539068963
+10 206 0.0001504098668872678
+10 220 0.0003008197337745356
+10 228 0.0030834022711889904
+10 246 0.003985861472512596
+10 247 0.0012784838685417762
+10 262 0.003910656539068963
+10 263 0.0011280740016545085
+10 269 0.0032338121380762574
+10 270 0.002857787470858088
+10 271 0.0003008197337745356
+10 276 0.000902459201323607
+10 277 0.00556516507482891
+10 288 0.0027825825374144545
+10 356 0.020305332029781156
+10 357 0.019703692562232082
+10 358 0.02549447243739189
+10 359 0.023764758968188315
+10 360 0.02587049710461006
+10 361 0.022486275099646538
+10 362 0.022411070166202904
+10 363 0.02278709483342107
+10 364 0.026321726705271865
+10 365 0.02007971722945025
+10 366 0.016093855756937656
+10 367 0.022260660299315636
+10 368 0.011882379484094157
+10 369 0.009400616680454237
+10 370 0.00962623148078514
+10 371 0.011431149883432353
+10 372 0.021583815898322933
+10 373 0.024742423102955553
+10 374 0.01947807776190118
+10 375 0.01789877415958487
+10 376 0.023388734300970146
+10 377 0.023689554034744677
+10 378 0.009400616680454237
+10 379 0.005865984808603443
+10 380 0.01135594494998872
+10 381 0.022486275099646538
+10 382 0.015341806422501316
+10 383 0.01135594494998872
+10 384 0.01158155975031962
+10 385 0.019703692562232082
+10 386 0.01504098668872678
+10 387 0.018124388959915774
+10 388 0.010077461081446944
+10 389 0.02293750470030834
+10 390 0.01383770775362864
+10 391 0.017372339625479433
+10 392 0.019703692562232086
+10 393 0.011882379484094157
+10 394 0.024667218169511923
+10 395 0.024667218169511916
+10 396 0.012333609084755958
+10 397 0.011506354816875987
+10 398 0.013236068286079568
+10 399 0.0070692637437015865
+10 400 0.01940287282845755
+10 401 0.016093855756937656
+10 402 0.020530946830112053
+10 403 0.008197337745356097
+10 404 0.01759795442581033
+10 405 0.021508610964879295
+10 406 0.008197337745356095
+10 407 0.013988117620515906
+10 408 0.008949387079792434
+10 409 0.006467624276152515
+10 410 0.005264345341054373
+10 411 0.005565165074828909
+10 412 0.003835451605625329
+10 413 0.002105738136421749
+10 414 0.0012784838685417764
+10 415 0.002556967737083553
+10 417 7.52049334436339e-05
+10 420 0.0020305332029781154
+10 421 0.0006016394675490712
+10 422 0.0006016394675490712
+10 427 7.52049334436339e-05
+10 430 0.004737910806948936
+10 431 0.002331352936752651
+10 432 0.0001504098668872678
+10 440 0.0010528690682108748
+10 441 0.0021057381364217496
+10 446 7.52049334436339e-05
+10 452 0.004512296006618034
+10 453 0.003609836805294428
+10 454 0.0006016394675490712
+10 456 0.0006016394675490712
+10 457 0.0004512296006618035
+11 356 0.011297349184080336
+11 357 0.011888060252528984
+11 358 0.004430333013364838
+11 359 0.004430333013364838
+11 360 0.009229860444510078
+11 361 0.011371188067636416
+11 362 0.0038396219449161927
+11 363 0.002805877575131064
+11 364 0.005759432917374288
+11 365 0.0014767776711216124
+11 366 0.0003691944177804031
+11 367 0.0014029387875655322
+11 372 0.011371188067636418
+11 373 0.004504171896920917
+11 374 0.0012552610204533705
+11 375 0.0011075832533412094
+11 376 0.005316399616037805
+11 377 0.005685594033818208
+11 381 0.001772133205345935
+11 382 0.0003691944177804031
+11 385 0.00118142213689729
+11 386 0.0005168721848925644
+11 387 0.0011075832533412094
+11 388 7.383888355608063e-05
+11 389 0.0031012331093553864
+11 390 0.019345787491693123
+11 391 0.010928154766299934
+11 392 0.01299564350587019
+11 393 0.02082256516281474
+11 394 0.0057594329173742895
+11 395 0.00945137709517832
+11 396 0.017352137635678947
+11 397 0.02001033744369785
+11 398 0.018238204238351912
+11 400 0.01794284870412759
+11 401 0.019124270841024884
+11 402 0.016170715498781657
+11 403 0.022816215018828915
+11 404 0.01727829875212287
+11 405 0.014546260060547885
+11 406 0.0239976371557262
+11 407 0.022963892785941076
+11 408 0.02695119249796943
+11 409 0.023776120505057962
+11 410 0.019493465258805284
+11 411 0.023849959388614037
+11 412 0.026581998080189025
+11 413 0.020601048512146496
+11 414 0.019493465258805288
+11 415 0.02163479288193162
+11 416 0.004873366314701322
+11 417 0.007900760540500627
+11 418 0.0042088163626965965
+11 419 0.0016982943217898545
+11 420 0.018238204238351912
+11 421 0.012035738019641142
+11 422 0.012331093553865465
+11 423 0.0055379162667060465
+11 424 0.004061138595584434
+11 425 0.0016982943217898542
+11 426 0.0008122277191168869
+11 427 0.00834379384183711
+11 428 0.0005168721848925643
+11 429 0.0015506165546776932
+11 430 0.023406926087277558
+11 431 0.019124270841024884
+11 432 0.016392232149449903
+11 433 0.005907110684486449
+11 434 0.0019198109724580966
+11 435 0.015432326663220851
+11 436 0.006940855054271579
+11 437 0.0013290999040094513
+11 438 0.013364837923650594
+11 439 0.00694085505427158
+11 440 0.02126559846415122
+11 441 0.02355460385438972
+11 442 0.002732038691574983
+11 444 7.383888355608063e-05
+11 446 0.010854315882743852
+11 447 0.0031012331093553864
+11 448 0.007753082773388465
+11 449 0.0018459720889020155
+11 450 0.00044303330133648377
+11 451 0.00044303330133648377
+11 452 0.023776120505057962
+11 453 0.02229934283393635
+11 454 0.02126559846415122
+11 455 0.013290999040094512
+11 456 0.018385882005464073
+11 457 0.015580004430333012
+11 458 0.010189765930739126
+11 459 0.012035738019641142
+11 460 0.0034704275271357893
+11 461 0.004578010780476998
+11 462 0.005907110684486449
+11 463 0.000590711068448645
+11 464 0.000590711068448645
+11 465 0.0002953555342243225
+11 466 0.0019936498560141768
+11 467 0.0013290999040094513
+12 444 1.0
+13 16 0.0014635288607891346
+13 17 0.002575810794988877
+13 18 0.005737033134293408
+13 19 0.001990399250673223
+13 20 0.007785973539398196
+13 21 0.008664090855871677
+13 22 0.002985598876009834
+13 23 0.002224563868399485
+13 63 5.854115443156538e-05
+13 66 0.0018147757873785268
+13 67 0.0006439526987472192
+13 68 0.0002927057721578269
+13 69 0.0008195761620419153
+13 70 0.0007024938531787846
+13 71 0.0033953869570307925
+13 72 0.0024001873316941806
+13 73 0.00023416461772626153
+13 74 0.009308043554618896
+13 75 0.007551808921671934
+13 76 0.01890879288139562
+13 77 0.013230300901533777
+13 80 0.0013464465519260039
+13 81 0.0002927057721578269
+13 82 0.0016976934785153963
+13 83 0.0040978808102095764
+13 93 0.00017562346329469617
+13 100 0.00017562346329469617
+13 102 0.00011708230886313077
+13 103 0.00035124692658939234
+13 137 0.00011708230886313077
+13 141 0.020021074815595362
+13 142 0.016625687858564567
+13 143 0.0016391523240838306
+13 144 0.0005268703898840885
+13 145 0.0002927057721578269
+13 146 0.002868516567146704
+13 147 0.006673691605198454
+13 148 0.008839714319166374
+13 149 0.0002927057721578269
+13 150 0.0002927057721578269
+13 151 0.0012293642430628731
+13 152 0.0011122819341997424
+13 157 0.0008781173164734808
+13 158 0.0004097880810209577
+13 160 0.02681184872965695
+13 161 0.023592085235920848
+13 162 0.03096827069429809
+13 163 0.02476290832455216
+13 178 0.0002927057721578269
+13 179 5.854115443156538e-05
+13 180 0.0009366584709050461
+13 181 0.00444912773679897
+13 182 0.013464465519260038
+13 183 0.0167427701674277
+13 184 0.00017562346329469617
+13 185 5.854115443156538e-05
+13 186 0.0002927057721578269
+13 187 0.0008195761620419153
+13 196 0.017503805175038047
+13 197 0.023416461772626154
+13 198 0.023416461772626154
+13 199 0.02921203606135113
+13 201 0.0018733169418100922
+13 202 0.006439526987472192
+13 206 0.015162158997775435
+13 207 0.0006439526987472192
+13 218 0.0007610350076103501
+13 219 0.00046832923545252306
+13 220 0.006673691605198454
+13 227 0.00011708230886313077
+13 228 0.0009951996253366115
+13 246 0.0106544901065449
+13 247 0.014576747453459781
+13 262 0.00011708230886313077
+13 268 0.0033368458025992264
+13 269 0.010420325488818641
+13 270 0.0035710104203254887
+13 271 0.002985598876009834
+13 275 0.009834913944502985
+13 276 0.02142606252195293
+13 277 0.01164968973188151
+13 278 0.00035124692658939234
+13 288 0.004741833508956796
+13 289 0.014693829762322912
+13 290 0.02207001522070015
+13 291 0.017913593256059006
+13 292 0.011005737033134292
+13 293 0.010478866643250203
+13 358 0.0003512469265893923
+13 363 5.854115443156538e-05
+13 365 0.00035124692658939234
+13 366 0.0007024938531787846
+13 367 0.00017562346329469617
+13 368 0.00017562346329469617
+13 369 0.0009951996253366115
+13 370 0.005151621589977754
+13 371 0.005385786207704015
+13 374 0.0016976934785153963
+13 375 0.0017562346329469615
+13 376 0.0004097880810209577
+13 377 0.0003512469265893923
+13 378 0.00046832923545252306
+13 379 0.0015220700152206996
+13 381 0.0014635288607891346
+13 382 0.0009951996253366115
+13 383 0.00532724505327245
+13 384 0.0037466338836201845
+13 386 0.0011708230886313077
+13 387 0.00011708230886313077
+13 388 0.0012293642430628731
+13 389 0.0002927057721578269
+13 394 0.00017562346329469617
+13 399 0.0033953869570307925
+13 468 5.854115443156538e-05
+13 469 0.0011122819341997424
+13 470 0.0027514342582835734
+13 471 0.0012879053974944384
+13 474 5.854115443156538e-05
+13 475 0.0002927057721578269
+13 476 0.0002927057721578269
+13 477 0.0018147757873785268
+13 478 0.0020489404051047887
+13 479 0.0011122819341997424
+13 480 0.004332045427935838
+13 481 0.006556609296335323
+13 483 0.00046832923545252306
+13 484 0.012352183585060298
+13 485 0.014869453225617611
+13 486 0.005912656597588104
+13 487 0.004214963119072708
+13 488 0.01164968973188151
+13 489 0.015806111696522657
+13 490 0.008312843929282283
+13 491 0.009834913944502985
+13 492 0.006146821215314366
+13 493 0.015513405924364829
+13 494 0.02007961597002693
+13 495 0.0024001873316941806
+13 496 0.008956796628029503
+13 497 0.004741833508956796
+13 498 0.003512469265893923
+13 499 0.002517269640557311
+13 501 0.0005854115443156538
+13 502 0.0004097880810209577
+13 504 0.001990399250673223
+13 505 0.00040978808102095764
+13 509 0.010478866643250205
+13 510 0.02207001522070015
+13 513 0.0012293642430628731
+13 579 0.021660227139679192
+13 580 0.0002927057721578269
+13 581 0.00011708230886313077
+13 582 0.0012879053974944388
+13 583 0.0018147757873785272
+13 584 5.854115443156538e-05
+13 585 0.00011708230886313077
+13 586 0.0011122819341997422
+13 587 0.0008195761620419154
+13 589 0.0007610350076103501
+13 590 0.003395386957030792
+13 591 0.0026928931038520073
+13 592 0.009834913944502985
+13 593 0.009834913944502985
+13 594 0.00011708230886313077
+13 595 0.0013464465519260039
+13 596 0.0015806111696522653
+13 597 0.0002927057721578269
+13 598 0.00023416461772626153
+13 599 0.0009951996253366115
+13 600 0.0002927057721578269
+13 601 0.0012293642430628731
+13 602 0.00046832923545252306
+13 603 0.00011708230886313077
+13 604 0.003980798501346446
+13 605 0.013523006673691603
+13 606 0.011591148577449948
+13 607 0.006263903524177495
+13 608 0.014693829762322912
+13 610 0.0003512469265893923
+13 611 0.0012293642430628734
+13 612 5.854115443156538e-05
+13 613 0.005327245053272449
+13 614 0.0019318580962416575
+13 615 0.006615150450766888
+13 616 0.0026928931038520073
+13 617 0.0002927057721578269
+13 627 0.005268703898840884
+13 630 0.00011708230886313077
+13 696 0.00023416461772626153
+13 769 0.00076103500761035
+13 770 0.004683292354525231
+13 771 0.0011122819341997424
+13 772 5.854115443156538e-05
+13 774 0.00076103500761035
+13 775 0.003512469265893923
+13 776 0.008020138157124457
+14 74 0.0005157677571470676
+14 75 0.0005157677571470676
+14 76 0.004273504273504274
+14 77 0.0008104921898025347
+14 141 0.002799882110226938
+14 142 0.0003684055408193339
+14 160 0.001326259946949602
+14 161 0.0005894488653109342
+14 162 0.004420866489832007
+14 163 0.0050103153551429415
+14 196 7.368110816386678e-05
+14 197 0.0014736221632773356
+14 198 0.0030209254347185383
+14 199 0.0009578544061302684
+14 206 7.368110816386678e-05
+14 246 0.0013262599469496023
+14 247 0.0061155319776009425
+14 269 7.368110816386678e-05
+14 276 0.0034630120837017397
+14 277 0.0008841732979664015
+14 290 0.001399941055113469
+14 291 0.0052313586796345415
+14 292 0.0058944886531093425
+14 293 0.008989095195991748
+14 468 0.0199675803124079
+14 469 0.02460949012673151
+14 470 0.021220159151193633
+14 471 0.02586206896551724
+14 472 0.020704391394046565
+14 473 0.017978190391983492
+14 474 0.020114942528735632
+14 475 0.02586206896551724
+14 476 0.02291482463896257
+14 477 0.02475685234305924
+14 478 0.021293840259357502
+14 479 0.026009431181844976
+14 480 0.019451812555260833
+14 481 0.014294134983790155
+14 482 0.01422045387562629
+14 483 0.02726201002063071
+14 484 0.02026230474506337
+14 485 0.015694076038903628
+14 486 0.02726201002063071
+14 487 0.02733569112879458
+14 488 0.01215738284703802
+14 489 0.009652225169466549
+14 490 0.015767757147067494
+14 491 0.02460949012673151
+14 492 0.020114942528735635
+14 493 0.013704686118479222
+14 494 0.01333628057765989
+14 495 0.022988505747126436
+14 496 0.018272914824638966
+14 497 0.020851753610374304
+14 498 0.016578249336870028
+14 499 0.025567344532861774
+14 500 0.007515473032714411
+14 501 0.019157088122605366
+14 502 0.015104627173592693
+14 503 0.00987326849395815
+14 504 0.021293840259357502
+14 505 0.020999115826702035
+14 506 0.013262599469496024
+14 507 0.013483642793987621
+14 508 0.010389036251105217
+14 509 0.011715296198054817
+14 510 0.010167992926613616
+14 511 0.011199528440907752
+14 512 0.009357500736811082
+14 513 0.020335985853227233
+14 514 0.010683760683760684
+14 515 0.01215738284703802
+14 516 0.016357206012378427
+14 517 0.004052460949012673
+14 518 0.006704980842911877
+14 519 0.004273504273504274
+14 520 0.0036103743000294726
+14 521 0.004494547597995874
+14 522 0.003020925434718538
+14 523 0.002136752136752137
+14 524 0.0037577365163572064
+14 525 0.0005894488653109342
+14 526 0.0008104921898025347
+14 531 0.0002947244326554671
+14 541 0.0016209843796050694
+14 542 0.0006631299734748011
+14 551 0.0019157088122605363
+14 552 0.0009578544061302684
+14 563 0.005010315355142941
+14 564 0.004715590922487474
+14 565 0.0010315355142941351
+14 567 0.000663129973474801
+14 568 0.00022104332449160037
+14 579 0.006115531977600943
+15 468 0.01103996467211305
+15 469 0.010230367262824759
+15 470 0.0023551924633841174
+15 471 0.004121586810922205
+15 472 0.009199970560094207
+15 473 0.011334363730036065
+15 474 0.004047987046441452
+15 475 0.0027967910502686394
+15 476 0.0059615809229410476
+15 477 0.0014719952896150733
+15 478 0.0003679988224037683
+15 479 0.0016191948185765807
+15 482 0.011187164201074557
+15 483 0.0056671818650180315
+15 484 0.0014719952896150733
+15 485 0.0003679988224037683
+15 486 0.004563185397806727
+15 487 0.0073599764480753675
+15 491 0.002134393169941856
+15 492 0.0003679988224037683
+15 495 0.0011775962316920587
+15 496 0.0005151983513652757
+15 497 0.0005151983513652757
+15 498 7.359976448075367e-05
+15 499 0.0032383896371531613
+15 500 0.019945536174284243
+15 501 0.01781114300434239
+15 502 0.014204754544785456
+15 503 0.02524471921689851
+15 504 0.005446382571575771
+15 505 0.010524766320747773
+15 506 0.01832634135570766
+15 507 0.01884153970707294
+15 508 0.018473540884669168
+15 511 0.01781114300434239
+15 512 0.019356738058438214
+15 513 0.012143961139324354
+15 514 0.020755133583572533
+15 515 0.01862074041363068
+15 516 0.015014351954073748
+15 517 0.024361522043129462
+15 518 0.02333112534039891
+15 519 0.027011113564436594
+15 520 0.02465592110105248
+15 521 0.024067122985206444
+15 522 0.024508721572090966
+15 523 0.023478324869360415
+15 524 0.025980716861706044
+15 525 0.018031942297784646
+15 526 0.020607934054611025
+15 527 0.0012511959961728123
+15 528 0.0059615809229410476
+15 529 0.0025023919923456246
+15 530 0.0009567969382497977
+15 531 0.018179141826746157
+15 532 0.006255979980864061
+15 533 0.011187164201074557
+15 534 0.005225583278133511
+15 535 0.004710384926768235
+15 536 0.0016927945830573343
+15 537 0.0007359976448075366
+15 538 0.013247957606535658
+15 540 7.359976448075367e-05
+15 541 0.02340472510487966
+15 542 0.02031353499668801
+15 543 0.010745565614190034
+15 544 0.0032383896371531613
+15 545 0.0003679988224037683
+15 546 0.015529550305439023
+15 547 0.005593582100537278
+15 548 0.001103996467211305
+15 549 0.019356738058438214
+15 550 0.009126370795613454
+15 551 0.025465518510340766
+15 552 0.022374328402149115
+15 553 0.0029439905792301465
+15 557 0.011923161845882095
+15 558 0.0029439905792301465
+15 559 0.00942076985353647
+15 560 0.003679988224037683
+15 561 0.0002943990579230147
+15 563 0.019356738058438214
+15 564 0.024582321336571723
+15 565 0.02244792816662987
+15 566 0.015382350776477514
+15 567 0.019503937587399718
+15 568 0.015161551483035255
+15 569 0.0059615809229410476
+15 570 0.01023036726282476
+15 571 0.0030175903437109006
+15 572 0.003459188930595423
+15 573 0.005519982336056524
+15 574 0.0008095974092882903
+15 575 0.0008095974092882903
+15 576 7.359976448075367e-05
+15 577 0.0013247957606535659
+15 578 0.0008095974092882903
+16 556 1.0
+17 17 0.0004919184820801125
+17 18 0.0006324666198172875
+17 20 0.005762473647224175
+17 21 0.0021082220660576245
+17 22 0.0014757554462403375
+17 23 0.0024595924104005625
+17 70 0.000140548137737175
+17 71 0.000983836964160225
+17 72 0.0023190442726633872
+17 73 0.0004919184820801125
+17 76 0.000140548137737175
+17 77 0.0006324666198172875
+17 80 0.008151791988756148
+17 81 0.006676036542515813
+17 82 0.016303583977512297
+17 83 0.012297962052002813
+17 96 0.002178496134926213
+17 97 0.0007027406886858749
+17 98 7.02740688685875e-05
+17 99 0.0011243851018974
+17 100 0.009065354884047786
+17 101 0.007308503162333099
+17 102 0.015038650737877725
+17 103 0.017919887561489812
+17 141 0.0018271257905832748
+17 142 0.003794799718903725
+17 148 0.0013352073085031624
+17 153 0.001546029515108925
+17 154 0.0024595924104005625
+17 160 0.0134926212227688
+17 161 0.01883345045678145
+17 162 0.012438510189739986
+17 163 0.005200281096275476
+17 178 0.0007730147575544624
+17 179 0.0007730147575544624
+17 180 0.003021784961349263
+17 181 0.00758959943780745
+17 182 0.01377371749824315
+17 183 0.007238229093464512
+17 184 0.0026001405481377374
+17 196 0.0123682361208714
+17 197 0.007449051300070275
+17 198 0.0071679550245959235
+17 199 0.0202389318341532
+17 200 0.0004919184820801125
+17 201 0.027406886858749122
+17 202 0.020028109627547436
+17 206 0.0019676739283204497
+17 207 0.0004919184820801125
+17 218 0.000140548137737175
+17 220 0.00028109627547435
+17 247 7.02740688685875e-05
+17 256 0.000140548137737175
+17 257 0.0004919184820801125
+17 269 0.00035137034434293746
+17 275 0.005270555165144062
+17 276 0.0010541110330288123
+17 277 7.02740688685875e-05
+17 278 0.02508784258608574
+17 289 0.019465917076598734
+17 290 0.0044975404075896
+17 291 0.001546029515108925
+17 292 0.0002108222066057625
+17 293 7.02740688685875e-05
+17 484 0.0009135628952916374
+17 485 0.0007730147575544624
+17 489 0.000421644413211525
+17 491 0.0004919184820801125
+17 492 7.02740688685875e-05
+17 493 0.0013352073085031622
+17 494 0.002951510892480675
+17 509 0.0002108222066057625
+17 510 0.0033731553056922
+17 579 0.0027406886858749122
+17 580 0.003162333099086437
+17 581 0.0023190442726633877
+17 582 0.009978917779339425
+17 583 0.009065354884047788
+17 584 0.0010541110330288125
+17 585 0.0016865776528460997
+17 586 0.004356992269852425
+17 587 0.003513703443429375
+17 588 0.0016865776528461
+17 589 0.004567814476458187
+17 590 0.009065354884047786
+17 591 0.006886858749121575
+17 592 0.013914265635980324
+17 593 0.016795502459592413
+17 594 0.021503865073787775
+17 595 0.028742094167252288
+17 596 0.02178496134926212
+17 597 0.02059030217849614
+17 598 0.0026001405481377374
+17 599 0.004427266338721012
+17 600 0.015038650737877721
+17 601 0.015390021082220663
+17 602 0.008081517919887564
+17 603 0.007308503162333099
+17 604 0.021995783555867888
+17 605 0.02312016865776529
+17 606 0.019465917076598734
+17 607 0.025720309205903027
+17 608 0.019465917076598737
+17 609 0.0123682361208714
+17 610 0.013070976809557275
+17 611 0.02009838369641603
+17 612 0.019184820801124384
+17 613 0.012719606465214337
+17 614 0.0134926212227688
+17 615 0.0179901616303584
+17 616 0.015319747013352071
+17 617 0.005621925509486999
+17 618 0.00028109627547435
+17 619 0.0022487702037948
+17 620 0.0026704146170063252
+17 621 0.0002108222066057625
+17 622 0.007589599437807451
+17 623 0.0018271257905832748
+17 624 0.00084328882642305
+17 625 0.0007027406886858749
+17 626 0.0009135628952916376
+17 627 0.01981728742094167
+17 628 0.0002108222066057625
+17 629 0.0004919184820801124
+17 630 0.003021784961349262
+17 631 0.0006324666198172875
+17 632 7.02740688685875e-05
+17 633 7.02740688685875e-05
+17 696 0.021152494729444835
+17 769 0.0179901616303584
+17 770 0.020941672522839076
+17 771 0.011243851018974
+17 772 0.005411103302881238
+17 773 0.0028109627547434997
+17 774 0.0036542515811665496
+17 775 0.00871398453970485
+17 776 0.012157413914265636
+17 777 0.00035137034434293746
+18 82 0.0006012777151446825
+18 83 0.00015031942878617063
+18 103 0.0004509582863585119
+18 142 7.515971439308531e-05
+18 160 0.001428034573468621
+18 161 0.0071401728673431055
+18 162 0.0018789928598271326
+18 182 0.0005261180007515971
+18 196 0.0018789928598271326
+18 197 0.0009019165727170237
+18 198 0.0008267568583239384
+18 199 0.006238256294626081
+18 201 0.003006388575723412
+18 202 0.0018038331454340473
+18 275 0.00015031942878617063
+18 278 0.002179631717399474
+18 289 0.003607666290868095
+18 580 0.020142803457346866
+18 581 0.01698609545283728
+18 582 0.03013904547162721
+18 583 0.02705749718151071
+18 584 0.017737692596768138
+18 585 0.016910935738444193
+18 586 0.021270199173243146
+18 587 0.018489289740698987
+18 588 0.01924088688462984
+18 589 0.016910935738444197
+18 590 0.016234498308906428
+18 591 0.017061255167230362
+18 592 0.015182262307403233
+18 593 0.010672679443818113
+18 594 0.004509582863585118
+18 595 0.005712138293874483
+18 596 0.01570838030815483
+18 597 0.011424276587748966
+18 598 0.018714768883878245
+18 599 0.01969184517098835
+18 600 0.020593761743705377
+18 601 0.023900789177001128
+18 602 0.027583615182262308
+18 603 0.0266065388951522
+18 604 0.007591131153701616
+18 605 0.007666290868094702
+18 606 0.012701991732431419
+18 607 0.020668921458098462
+18 608 0.016309658023299513
+18 609 0.015031942878617064
+18 610 0.017061255167230362
+18 611 0.009845922585494176
+18 612 0.009845922585494176
+18 613 0.02435174746335964
+18 614 0.02450206689214581
+18 615 0.02720781661029688
+18 616 0.016459977452085682
+18 617 0.021119879744456973
+18 618 0.012251033446072902
+18 619 0.020744081172491546
+18 620 0.02247275460353251
+18 621 0.008643367155204812
+18 622 0.025779782036828264
+18 623 0.02006764374295378
+18 624 0.011724915445321307
+18 625 0.01089815858699737
+18 626 0.011800075159714395
+18 627 0.006914693724163849
+18 628 0.011499436302142051
+18 629 0.011273957158962795
+18 630 0.0214956783164224
+18 631 0.012251033446072907
+18 632 0.013077790304396842
+18 633 0.014656144306651634
+18 634 0.00496054114994363
+18 635 0.006914693724163849
+18 636 0.004735062006764375
+18 637 0.004359263434798948
+18 638 0.004208944006012777
+18 639 0.003757985719654265
+18 640 0.0021796317173994736
+18 641 0.003908305148440436
+18 642 0.0010522360015031943
+18 643 0.0006012777151446825
+18 648 0.0005261180007515971
+18 658 0.0020293122886133035
+18 659 0.0011273957158962795
+18 668 0.000751597143930853
+18 669 0.0011273957158962795
+18 680 0.002931228861330327
+18 681 0.0057872980082675695
+18 682 0.001428034573468621
+18 684 0.0006012777151446825
+18 685 0.0006012777151446825
+18 696 0.003757985719654265
+18 769 0.0009019165727170238
+18 770 0.002931228861330327
+18 771 0.00015031942878617063
+18 775 0.00022547914317925594
+18 776 0.001202555430289365
+19 580 0.012027744982290436
+19 581 0.009961629279811098
+19 582 0.0059031877213695395
+19 583 0.004574970484061393
+19 584 0.009223730814639905
+19 585 0.011289846517119244
+19 586 0.0038370720188902006
+19 587 0.0028040141676505316
+19 588 0.0057556080283353
+19 589 0.0014757969303423849
+19 590 0.0003689492325855962
+19 591 0.0008116883116883117
+19 598 0.011806375442739079
+19 599 0.004501180637544274
+19 600 0.0012544273907910272
+19 601 0.0012544273907910272
+19 602 0.005165289256198347
+19 603 0.005165289256198347
+19 609 0.0005165289256198347
+19 610 0.0008116883116883117
+19 613 0.0016971664698937428
+19 614 0.0010330578512396697
+19 615 0.0011068476977567888
+19 616 0.0003689492325855962
+19 617 0.0028040141676505316
+19 618 0.021325265643447465
+19 619 0.012322904368358915
+19 620 0.016528925619834708
+19 621 0.022432113341204252
+19 622 0.0042060212514757975
+19 623 0.009518890200708384
+19 624 0.021989374262101534
+19 625 0.015053128689492327
+19 626 0.01977567886658796
+19 628 0.017635773317591502
+19 629 0.019406729634002362
+19 630 0.010256788665879575
+19 631 0.01977567886658796
+19 632 0.018299881936245575
+19 633 0.014684179456906728
+19 634 0.027007083825265645
+19 635 0.02317001180637544
+19 636 0.026638134592680048
+19 637 0.02368654073199528
+19 638 0.02435064935064935
+19 639 0.02457201889020071
+19 640 0.021915584415584416
+19 641 0.0256788665879575
+19 642 0.018816410861865408
+19 643 0.01682408500590319
+19 644 0.002656434474616293
+19 645 0.009445100354191263
+19 646 0.0031729634002361272
+19 647 0.0025826446280991736
+19 648 0.016602715466351833
+19 649 0.006419716646989375
+19 650 0.010478158205430934
+19 651 0.004870129870129871
+19 652 0.003246753246753247
+19 653 0.0014757969303423849
+19 654 0.0013282172373081465
+19 655 0.010847107438016527
+19 656 0.0005903187721369539
+19 657 0.0005165289256198347
+19 658 0.02169421487603306
+19 659 0.019406729634002362
+19 660 0.007747933884297522
+19 661 0.001844746162927981
+19 662 0.00014757969303423848
+19 663 0.012101534828807558
+19 664 0.0038370720188902014
+19 665 0.0007378984651711924
+19 666 0.0157172373081464
+19 667 0.006050767414403779
+19 668 0.02221074380165289
+19 669 0.021103896103896107
+19 670 0.0028040141676505316
+19 674 0.006419716646989373
+19 675 0.0014020070838252656
+19 676 0.010109208972845335
+19 677 0.0030253837072018895
+19 678 7.378984651711924e-05
+19 679 0.00014757969303423848
+19 680 0.02206316410861866
+19 681 0.022432113341204252
+19 682 0.0256788665879575
+19 683 0.01977567886658796
+19 684 0.022358323494687134
+19 685 0.02088252656434475
+19 686 0.012470484061393153
+19 687 0.01586481700118064
+19 688 0.004427390791027155
+19 689 0.006198347107438017
+19 690 0.00974025974025974
+19 691 0.000885478158205431
+19 692 0.0003689492325855962
+19 693 0.00014757969303423848
+19 694 0.002877804014167651
+19 695 0.0016233766233766235
+20 673 1.0

data/body_models/SMPLX_to_J14.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5df844ddea85b0a400a2e8dbe63d09d19f2b1b7ec0e0e952daeae08f83d82d61
+size 4692193

data/body_models/SMPL_NEUTRAL.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4924f235e63f7c5d5b690acedf736419c2edb846a2d69fc0956169615fa75688
+size 247186228

data/body_models/all_means.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:010c2178eff5fd58d07bab3717002e959fe62541aaaef778b09414ec0237690d
+size 4758

data/body_models/downsample_mat_smplx.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b67d12e8e9af767d9856fea8cb3366bfa8025fdf17cd4e25fc8b10f9a45eca9e
+size 18310685

data/body_models/joints_regressor_cmr.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a408b885040d714c94b41f64b2ec329d20dce673ae330d04a07a4b02dae7a13d
+size 661568

data/body_models/smpl/SMPL_FEMALE.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a583c1b98e4afc19042641f1bae5cd8a1f712a6724886291a7627ec07acd408d
+size 39056454

data/body_models/smpl/SMPL_MALE.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e8c0bbbbc635dcb166ed29c303fb4bef16ea5f623e5a89263495a9e403575bd
+size 39056404

data/body_models/smpl/SMPL_NEUTRAL.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4924f235e63f7c5d5b690acedf736419c2edb846a2d69fc0956169615fa75688
+size 247186228

data/body_models/smpl/index.html

@@ -0,0 +1,17 @@
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<html>
+<head>
+<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
+<title>Directory listing for /body_models/smpl/</title>
+</head>
+<body>
+<h1>Directory listing for /body_models/smpl/</h1>
+<hr>
+<ul>
+<li><a href="SMPL_FEMALE.pkl">SMPL_FEMALE.pkl</a></li>
+<li><a href="SMPL_MALE.pkl">SMPL_MALE.pkl</a></li>
+<li><a href="SMPL_NEUTRAL.pkl">SMPL_NEUTRAL.pkl</a></li>
+</ul>
+<hr>
+</body>
+</html>

data/body_models/smpl_mean_params.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6fd6dd687800da946d0a0492383f973b92ec20f166a0b829775882868c35fcdd
+size 1310

data/body_models/smplx/MANO_SMPLX_vertex_ids.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e5abe70b6574de25470475091e8008314a5b90127eb48c3e63bfa0adf8c04dcf
+size 13535

data/body_models/smplx/SMPL-X__FLAME_vertex_ids.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7e70cdc3659aae699b9732e8dd4af49106310c69b90dc83d9f73e96dbf871e49
+size 40312

data/body_models/smplx/SMPLX_FEMALE.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2a3686c9d6d218ff6822fba411c607a3c8125a70af340f384ce68bebecabe0e
+size 108794146

data/body_models/smplx/SMPLX_FEMALE.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3ac7af258fd217ab480b839c011545e5826cfa333ab34b3c98244ee3039bddd
+size 544434140

data/body_models/smplx/SMPLX_MALE.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab318e3f37d2bfaae26abf4e6fab445c2a610e1d63714794d60379cc263bc2a5
+size 108753445

data/body_models/smplx/SMPLX_MALE.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af7ebc82e44cf098598685474c0592049ddfaca8e850feb0c2b88343f9aacee3
+size 544477159

data/body_models/smplx/SMPLX_NEUTRAL.npz

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:376021446ddc86e99acacd795182bbef903e61d33b76b9d8b359c2b0865bd992
+size 108752058

data/body_models/smplx/SMPLX_NEUTRAL.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:381c808965deb4f5e845f8c3eddb0cd69930cc72e5774ce4f34c4ce3cf058361
+size 544173380

data/body_models/smplx/SMPLX_to_J14.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:be01f37aa99e794ace8f52abe7b31df302fe54c68e75062ea0431a6c2f5e084f
+size 1173328

data/body_models/smplx/SMPLX_to_J14.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5df844ddea85b0a400a2e8dbe63d09d19f2b1b7ec0e0e952daeae08f83d82d61
+size 4692193

data/body_models/smplx/smplx_kid_template.npy

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdce4f5886b9ddcb6da3ee0f70ae636b1aa1292f2b379c4c3149fce8abc0a604
+size 251528

data/body_models/smplx2smpl.pkl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1d912d121ad98132e4492d8e7a0f1a8cf4412811e14a7ef8cb337bb48eef99e
+size 578019251

datasets/AGORA_MM.py

@@ -0,0 +1,974 @@
+import os
+import os.path as osp
+from glob import glob
+import numpy as np
+from config.config import cfg
+import copy
+import json
+import pickle
+import cv2
+import torch
+from pycocotools.coco import COCO
+from util.human_models import smpl_x
+from util.preprocessing import load_img, sanitize_bbox, process_bbox, load_ply, load_obj
+from util.transforms import rigid_align, rigid_align_batch
+import tqdm
+import random
+from util.formatting import DefaultFormatBundle
+from detrsmpl.data.datasets.pipelines.transforms import Normalize
+import time
+from util.preprocessing import (
+    load_img, process_bbox, augmentation_instance_sample
+    ,process_human_model_output_batch_simplify,process_db_coord_batch_no_valid)
+# from util.human_models import smpl_x
+from .humandata import HumanDataset
+import csv
+KPS2D_KEYS = [
+    'keypoints2d_ori', 'keypoints2d_smplx', 'keypoints2d_smpl',
+    'keypoints2d_original','keypoints2d_gta'
+]
+KPS3D_KEYS = [
+    'keypoints3d_cam', 'keypoints3d', 'keypoints3d_smplx', 'keypoints3d_smpl',
+    'keypoints3d_original', 'keypoints3d_gta'
+]
+class AGORA_MM(HumanDataset):
+    def __init__(self, transform, data_split):
+        super(AGORA_MM, self).__init__(transform, data_split)
+        self.img_shape = [2160,3840]
+        pre_prc_file_train = 'spec_train_smpl.npz'
+        pre_prc_file_test = 'spec_test_smpl.npz'
+        self.save_idx = 0
+        if self.data_split == 'train':
+            filename = getattr(cfg, 'filename', pre_prc_file_train)
+        else:
+            self.test_set = 'val'
+        
+        self.img_dir = './data/datasets/agora'
+
+
+        if data_split == 'train':
+            if self.img_shape == [2160,3840]:
+                self.annot_path = 'data/preprocessed_npz/multihuman_data/agora_train_3840_w_occ_multi_2010.npz'
+                self.annot_path_cache = 'data/preprocessed_npz/cache/agora_train_3840_w_occ_cache_2010.npz'
+            elif self.img_shape == [720,1280]:
+                self.annot_path = 'data/preprocessed_npz/multihuman_data/agora_train_1280_multi_1010.npz'
+                self.annot_path_cache = 'data/preprocessed_npz/cache/agora_train_cache_1280_1010.npz'
+
+        elif data_split == 'test':
+            if self.img_shape == [2160,3840]:
+                self.annot_path = 'data/preprocessed_npz/multihuman_data/agora_validation_multi_3840_1010.npz'
+                self.annot_path_cache = 'data/preprocessed_npz/cache/agora_validation_cache_3840_1010_occ_cache_balance.npz'
+            elif self.img_shape == [720,1280]:
+                self.annot_path = 'data/preprocessed_npz/multihuman_data/agora_validation_1280_1010_occ.npz'
+                self.annot_path_cache = 'data/preprocessed_npz/cache/agora_validation_cache_1280_1010_occ.npz'
+        
+        self.use_cache = getattr(cfg, 'use_cache', False)
+        self.cam_param = {}
+
+        # load data or cache
+        if self.use_cache and osp.isfile(self.annot_path_cache):
+            print(f'[{self.__class__.__name__}] loading cache from {self.annot_path_cache}')
+            self.datalist = self.load_cache(self.annot_path_cache)
+        else:
+            if self.use_cache:
+                print(f'[{self.__class__.__name__}] Cache not found, generating cache...')
+            self.datalist = self.load_data(
+                train_sample_interval=getattr(cfg, f'{self.__class__.__name__}_train_sample_interval', 1))
+            if self.use_cache:
+                self.save_cache(self.annot_path_cache, self.datalist)
+
+
+    def load_data(self, train_sample_interval=1):
+
+        content = np.load(self.annot_path, allow_pickle=True)
+        
+        try:
+            frame_range = content['frame_range']
+        except KeyError:
+            frame_range = \
+                np.array([[i, i + 1] for i in range(self.num_data)])
+
+        num_examples = len(frame_range)
+        
+        if 'meta' in content:
+            meta = content['meta'].item()
+            print('meta keys:', meta.keys())
+        else:
+            meta = None
+            print(
+                'No meta info provided! Please give height and width manually')
+
+        print(
+            f'Start loading humandata {self.annot_path} into memory...\nDataset includes: {content.files}'
+        )
+        tic = time.time()
+        image_path = content['image_path']
+
+        if meta is not None and 'height' in meta:
+            height = np.array(meta['height'])
+            width = np.array(meta['width'])
+            image_shape = np.stack([height, width], axis=-1)
+        else:
+            image_shape = None
+
+        if meta is not None and 'gender' in meta and len(meta['gender']) != 0:
+            gender = meta['gender']
+        else:
+            gender = None
+        
+        if meta is not None and 'is_kid' in meta and len(meta['is_kid']) != 0:
+            is_kid = meta['is_kid']
+        else:
+            is_kid = None
+                    
+        bbox_xywh = content['bbox_xywh']
+
+        if 'smplx' in content:
+            smplx = content['smplx'].item()
+            as_smplx = 'smplx'
+        elif 'smpl' in content:
+            smplx = content['smpl'].item()
+            as_smplx = 'smpl'
+        elif 'smplh' in content:
+            smplx = content['smplh'].item()
+            as_smplx = 'smplh'
+        # TODO: temp solution, should be more general. But SHAPY is very special
+        elif self.__class__.__name__ == 'SHAPY':
+            smplx = {}
+        else:
+            raise KeyError('No SMPL for SMPLX available, please check keys:\n'
+                           f'{content.files}')
+
+        print('Smplx param', smplx.keys())
+
+        if 'lhand_bbox_xywh' in content and 'rhand_bbox_xywh' in content:
+            lhand_bbox_xywh = content['lhand_bbox_xywh']
+            rhand_bbox_xywh = content['rhand_bbox_xywh']
+        else:
+            lhand_bbox_xywh = np.zeros_like(bbox_xywh)
+            rhand_bbox_xywh = np.zeros_like(bbox_xywh)
+
+        if 'face_bbox_xywh' in content:
+            face_bbox_xywh = content['face_bbox_xywh']
+        else:
+            face_bbox_xywh = np.zeros_like(bbox_xywh)
+
+        decompressed = False
+        if content['__keypoints_compressed__']:
+            decompressed_kps = self.decompress_keypoints(content)
+            decompressed = True
+
+        keypoints3d = None
+        valid_kps3d = False
+        keypoints3d_mask = None
+        valid_kps3d_mask = False
+        
+        
+        # processing keypoints
+        for kps3d_key in KPS3D_KEYS:
+            if kps3d_key in content:
+                keypoints3d = decompressed_kps[kps3d_key][:, self.SMPLX_137_MAPPING, :] if decompressed \
+                else content[kps3d_key][:, self.SMPLX_137_MAPPING, :]
+                valid_kps3d = True
+                if keypoints3d.shape[-1] == 4:
+                    valid_kps3d_mask = True
+                break
+        if self.keypoints2d is not None:
+            keypoints2d = decompressed_kps[self.keypoints2d][:, self.SMPLX_137_MAPPING, :] if decompressed \
+                else content[self.keypoints2d][:, self.SMPLX_137_MAPPING, :]
+            
+
+        else:
+            for kps2d_key in KPS2D_KEYS:
+                if kps2d_key in content:
+                    keypoints2d = decompressed_kps[kps2d_key][:, self.SMPLX_137_MAPPING, :] if decompressed \
+                        else content[kps2d_key][:, self.SMPLX_137_MAPPING, :]
+                    
+        if keypoints2d.shape[-1] == 3:
+            valid_kps3d_mask = True
+        occlusion = content['meta'][()]['occ'] if 'occ' in content['meta'][()] and len(content['meta'][()]['occ'])>0 else None
+        
+        print('Done. Time: {:.2f}s'.format(time.time() - tic))
+
+        datalist = []
+        # num_examples
+
+        # processing each image, filter according to bbox valid
+        for i in tqdm.tqdm(range(int(num_examples))):
+            if self.data_split == 'train' and i % train_sample_interval != 0:
+                continue
+            frame_start, frame_end = frame_range[i]
+            img_path = osp.join(self.img_dir, image_path[frame_start])
+            # im_shape = cv2.imread(img_path).shape[:2]
+            img_shape = image_shape[
+                frame_start] if image_shape is not None else self.img_shape
+            
+
+            bbox_list = bbox_xywh[frame_start:frame_end, :4]
+            
+            valid_idx = []
+            body_bbox_list = []
+            
+            if hasattr(cfg, 'bbox_ratio'):
+                bbox_ratio = cfg.bbox_ratio * 0.833  # preprocess body bbox is giving 1.2 box padding
+            else:
+                bbox_ratio = 1.25
+            
+            for bbox_i, bbox in enumerate(bbox_list):
+                
+                bbox = process_bbox(bbox,
+                                    img_width=img_shape[1],
+                                    img_height=img_shape[0],
+                                    ratio=bbox_ratio)
+                if bbox is None:
+                    continue
+                else:
+                    valid_idx.append(frame_start + bbox_i)
+                    bbox[2:] += bbox[:2]
+                    body_bbox_list.append(bbox)
+            if len(valid_idx) == 0:
+                continue
+            valid_num = len(valid_idx)
+            # hand/face bbox
+            lhand_bbox_list = []
+            rhand_bbox_list = []
+            face_bbox_list = []
+            
+            for bbox_i in valid_idx:
+                lhand_bbox = lhand_bbox_xywh[bbox_i]
+                
+                rhand_bbox = rhand_bbox_xywh[bbox_i]
+                face_bbox = face_bbox_xywh[bbox_i]
+                if lhand_bbox[-1] > 0:  # conf > 0
+                    lhand_bbox = lhand_bbox[:4]
+                    if hasattr(cfg, 'bbox_ratio'):
+                        lhand_bbox = process_bbox(lhand_bbox,
+                                                  img_width=img_shape[1],
+                                                  img_height=img_shape[0],
+                                                  ratio=bbox_ratio)
+                    if lhand_bbox is not None:
+                        lhand_bbox[2:] += lhand_bbox[:2]  # xywh -> xyxy
+                else:
+                    lhand_bbox = None
+                if rhand_bbox[-1] > 0:
+                    rhand_bbox = rhand_bbox[:4]
+                    if hasattr(cfg, 'bbox_ratio'):
+                        rhand_bbox = process_bbox(rhand_bbox,
+                                                  img_width=img_shape[1],
+                                                  img_height=img_shape[0],
+                                                  ratio=bbox_ratio)
+                    if rhand_bbox is not None:
+                        rhand_bbox[2:] += rhand_bbox[:2]  # xywh -> xyxy
+                else:
+                    rhand_bbox = None
+                if face_bbox[-1] > 0:
+                    face_bbox = face_bbox[:4]
+                    if hasattr(cfg, 'bbox_ratio'):
+                        face_bbox = process_bbox(face_bbox,
+                                                 img_width=img_shape[1],
+                                                 img_height=img_shape[0],
+                                                 ratio=bbox_ratio)
+                    if face_bbox is not None:
+                        face_bbox[2:] += face_bbox[:2]  # xywh -> xyxy
+                else:
+                    face_bbox = None
+                lhand_bbox_list.append(lhand_bbox)
+                rhand_bbox_list.append(rhand_bbox)
+                face_bbox_list.append(face_bbox)
+            
+            # lhand_bbox = np.stack(lhand_bbox_list,axis=0)
+            # rhand_bbox = np.stack(rhand_bbox_list,axis=0)
+            # face_bbox = np.stack(face_bbox_list,axis=0)
+            joint_img = keypoints2d[valid_idx]
+            
+            # num_joints = joint_cam.shape[0]
+            # joint_valid = np.ones((num_joints, 1))
+            if valid_kps3d:
+                joint_cam = keypoints3d[valid_idx]
+            else:
+                joint_cam = None
+            
+            if 'leye_pose_0' in smplx.keys():
+                smplx.pop('leye_pose_0')
+            if 'leye_pose_1' in smplx.keys():
+                smplx.pop('leye_pose_1')
+            if 'leye_pose' in smplx.keys():
+                smplx.pop('leye_pose')
+            if 'reye_pose_0' in smplx.keys():
+                smplx.pop('reye_pose_0')
+            if 'reye_pose_1' in smplx.keys():
+                smplx.pop('reye_pose_1')
+            if 'reye_pose' in smplx.keys():
+                smplx.pop('reye_pose')
+            
+            occlusion_frame = occlusion[valid_idx] \
+                if occlusion is not None else np.array([1]*(valid_num))
+
+            smplx_param = {k: v[valid_idx] for k, v in smplx.items()}
+            gender_ = gender[valid_idx] \
+                if gender is not None else np.array(['neutral']*(valid_num))
+                
+            is_kid_ = is_kid[valid_idx] \
+                if is_kid is not None else np.array([1]*(valid_num))
+            lhand_bbox_valid = lhand_bbox_xywh[valid_idx,4]
+            rhand_bbox_valid = rhand_bbox_xywh[valid_idx,4]
+            face_bbox_valid = face_bbox_xywh[valid_idx,4]
+            
+            smplx_param['root_pose'] = smplx_param.pop('global_orient', None)
+            smplx_param['shape'] = smplx_param.pop('betas', None)
+            smplx_param['trans'] = smplx_param.pop('transl', np.zeros(3))
+            smplx_param['lhand_pose'] = smplx_param.pop('left_hand_pose', None)
+            smplx_param['rhand_pose'] = smplx_param.pop(
+                'right_hand_pose', None)
+            smplx_param['expr'] = smplx_param.pop('expression', None)
+
+            # TODO do not fix betas, give up shape supervision
+            if 'betas_neutral' in smplx_param and self.data_split == 'train':
+                smplx_param['shape'] = smplx_param.pop('betas_neutral')
+                # smplx_param['shape'] = np.zeros(10, dtype=np.float32)
+
+            if smplx_param['lhand_pose'] is None or self.body_only == True:
+                smplx_param['lhand_valid'] = np.zeros(valid_num, dtype=np.bool8)
+            else:
+                smplx_param['lhand_valid'] = lhand_bbox_valid.astype(np.bool8)
+                
+            if smplx_param['rhand_pose'] is None or self.body_only == True:
+                smplx_param['rhand_valid'] = np.zeros(valid_num, dtype=np.bool8)
+            else:
+                smplx_param['rhand_valid'] = rhand_bbox_valid.astype(np.bool8)
+            
+            if smplx_param['expr'] is None or self.body_only == True:
+                smplx_param['face_valid'] = np.zeros(valid_num, dtype=np.bool8)
+            else:
+                smplx_param['face_valid'] = face_bbox_valid.astype(np.bool8)
+
+            if joint_cam is not None and np.any(np.isnan(joint_cam)):
+                continue
+            
+            
+            datalist.append({
+                'img_path': img_path,
+                'img_shape': img_shape,
+                'bbox': body_bbox_list,
+                'lhand_bbox': lhand_bbox_list,
+                'rhand_bbox': rhand_bbox_list,
+                'face_bbox': face_bbox_list,
+                'joint_img': joint_img,
+                'joint_cam': joint_cam,
+                'smplx_param': smplx_param,
+                'as_smplx': as_smplx,
+                'gender': gender_,
+                'occlusion': occlusion_frame,
+                'is_kid': is_kid_,
+            })
+
+        # save memory
+        del content, image_path, bbox_xywh, lhand_bbox_xywh, rhand_bbox_xywh, face_bbox_xywh, keypoints3d, keypoints2d
+
+        if self.data_split == 'train':
+            print(f'[{self.__class__.__name__} train] original size:',
+                  int(num_examples), '. Sample interval:',
+                  train_sample_interval, '. Sampled size:', len(datalist))
+
+        if getattr(cfg, 'data_strategy',
+                   None) == 'balance' and self.data_split == 'train':
+            print(
+                f'[{self.__class__.__name__}] Using [balance] strategy with datalist shuffled...'
+            )
+            random.shuffle(datalist)
+
+        return datalist
+   
+    def __getitem__(self, idx):
+        try:
+            data = copy.deepcopy(self.datalist[idx])
+        except Exception as e:
+            print(f'[{self.__class__.__name__}] Error loading data {idx}')
+            print(e)
+            exit(0)
+
+        img_path, img_shape, bbox = \
+            data['img_path'], data['img_shape'], data['bbox']
+        as_smplx = data['as_smplx']
+        gender = data['gender'].copy()
+        for gender_str, gender_num in {
+            'neutral': -1, 'male': 0, 'female': 1}.items():
+            gender[gender==gender_str]=gender_num
+        gender = gender.astype(int)    
+        
+        img_whole_bbox = np.array([0, 0, img_shape[1], img_shape[0]])        
+        img = load_img(img_path, order='BGR')
+
+        num_person = len(data['bbox'])
+        data_name = self.__class__.__name__
+        img, img2bb_trans, bb2img_trans, rot, do_flip = \
+            augmentation_instance_sample(img, img_whole_bbox, self.data_split,data,data_name)
+        cropped_img_shape=img.shape[:2]
+        
+        num_person = len(data['bbox'])
+        if self.data_split == 'train':
+            joint_cam = data['joint_cam']  # num, 137,4
+            if joint_cam is not None:
+                dummy_cord = False
+                joint_cam[:,:,:3] = \
+                    joint_cam[:,:,:3] - joint_cam[:, self.joint_set['root_joint_idx'], None, :3]  # root-relative
+            else:
+                # dummy cord as joint_cam
+                dummy_cord = True
+                joint_cam = np.zeros(
+                    (num_person, self.joint_set['joint_num'], 4),
+                    dtype=np.float32)
+
+            joint_img = data['joint_img']
+            # do rotation on keypoints
+            joint_img_aug, joint_cam_wo_ra, joint_cam_ra, joint_trunc = \
+                process_db_coord_batch_no_valid(
+                    joint_img, joint_cam, do_flip, img_shape,
+                    self.joint_set['flip_pairs'], img2bb_trans, rot,
+                    self.joint_set['joints_name'], smpl_x.joints_name,
+                    cropped_img_shape)
+            joint_img_aug[:,:,2:] = joint_img_aug[:,:,2:] * joint_trunc
+            
+            # smplx coordinates and parameters
+            smplx_param = data['smplx_param']
+            smplx_pose, smplx_shape, smplx_expr, smplx_pose_valid, \
+            smplx_joint_valid, smplx_expr_valid, smplx_shape_valid = \
+                process_human_model_output_batch_simplify(
+                    smplx_param, do_flip, rot, as_smplx)
+            # if cam not provided, we take joint_img as smplx joint 2d, 
+            # which is commonly the case for our processed humandata
+            # change smplx_shape if use_betas_neutral
+            # processing follows that in process_human_model_output
+            
+            if self.use_betas_neutral:
+                smplx_shape = smplx_param['betas_neutral'].reshape(
+                    num_person, -1)
+                smplx_shape[(np.abs(smplx_shape) > 3).any(axis=1)] = 0.
+                smplx_shape = smplx_shape.reshape(num_person, -1)
+            # SMPLX joint coordinate validity
+            # for name in ('L_Big_toe', 'L_Small_toe', 'L_Heel', 'R_Big_toe', 'R_Small_toe', 'R_Heel'):
+            #     smplx_joint_valid[smpl_x.joints_name.index(name)] = 0
+            smplx_joint_valid = smplx_joint_valid[:, :, None]
+
+            lhand_bbox_center_list = []
+            lhand_bbox_valid_list = []
+            lhand_bbox_size_list = []
+            lhand_bbox_list = []
+            face_bbox_center_list = []
+            face_bbox_size_list = []
+            face_bbox_valid_list = []
+            face_bbox_list = []
+            rhand_bbox_center_list = []
+            rhand_bbox_valid_list = []
+            rhand_bbox_size_list = []
+            rhand_bbox_list = []
+            body_bbox_center_list = []
+            body_bbox_size_list = []
+            body_bbox_valid_list = []
+            body_bbox_list = []
+
+            for i in range(num_person):
+                body_bbox, body_bbox_valid = self.process_hand_face_bbox(
+                    data['bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                
+                lhand_bbox, lhand_bbox_valid = self.process_hand_face_bbox(
+                    data['lhand_bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                lhand_bbox_valid *= smplx_param['lhand_valid'][i]
+                
+                rhand_bbox, rhand_bbox_valid = self.process_hand_face_bbox(
+                    data['rhand_bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                rhand_bbox_valid *= smplx_param['rhand_valid'][i]
+                
+                face_bbox, face_bbox_valid = self.process_hand_face_bbox(
+                    data['face_bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                face_bbox_valid *= smplx_param['face_valid'][i]
+                
+                if do_flip:
+                    lhand_bbox, rhand_bbox = rhand_bbox, lhand_bbox
+                    lhand_bbox_valid, rhand_bbox_valid = rhand_bbox_valid, lhand_bbox_valid
+                    
+                body_bbox_list.append(body_bbox)
+                lhand_bbox_list.append(lhand_bbox)
+                rhand_bbox_list.append(rhand_bbox)
+                face_bbox_list.append(face_bbox)
+                
+                lhand_bbox_center = (lhand_bbox[0] + lhand_bbox[1]) / 2.
+                rhand_bbox_center = (rhand_bbox[0] + rhand_bbox[1]) / 2.
+                face_bbox_center = (face_bbox[0] + face_bbox[1]) / 2.
+                body_bbox_center = (body_bbox[0] + body_bbox[1]) / 2.
+                lhand_bbox_size = lhand_bbox[1] - lhand_bbox[0]
+                rhand_bbox_size = rhand_bbox[1] - rhand_bbox[0]
+
+                face_bbox_size = face_bbox[1] - face_bbox[0]
+                body_bbox_size = body_bbox[1] - body_bbox[0]
+                lhand_bbox_center_list.append(lhand_bbox_center)
+                lhand_bbox_valid_list.append(lhand_bbox_valid)
+                lhand_bbox_size_list.append(lhand_bbox_size)
+                face_bbox_center_list.append(face_bbox_center)
+                face_bbox_size_list.append(face_bbox_size)
+                face_bbox_valid_list.append(face_bbox_valid)
+                rhand_bbox_center_list.append(rhand_bbox_center)
+                rhand_bbox_valid_list.append(rhand_bbox_valid)
+                rhand_bbox_size_list.append(rhand_bbox_size)
+                body_bbox_center_list.append(body_bbox_center)
+                body_bbox_size_list.append(body_bbox_size)
+                body_bbox_valid_list.append(body_bbox_valid)
+            
+            
+            body_bbox = np.stack(body_bbox_list, axis=0)
+            lhand_bbox = np.stack(lhand_bbox_list, axis=0)
+            rhand_bbox = np.stack(rhand_bbox_list, axis=0)
+            face_bbox = np.stack(face_bbox_list, axis=0)
+            lhand_bbox_center = np.stack(lhand_bbox_center_list, axis=0)
+            lhand_bbox_valid = np.stack(lhand_bbox_valid_list, axis=0)
+            lhand_bbox_size = np.stack(lhand_bbox_size_list, axis=0)
+            face_bbox_center = np.stack(face_bbox_center_list, axis=0)
+            face_bbox_size = np.stack(face_bbox_size_list, axis=0)
+            face_bbox_valid = np.stack(face_bbox_valid_list, axis=0)
+            body_bbox_center = np.stack(body_bbox_center_list, axis=0)
+            body_bbox_size = np.stack(body_bbox_size_list, axis=0)
+            body_bbox_valid = np.stack(body_bbox_valid_list, axis=0)
+            rhand_bbox_center = np.stack(rhand_bbox_center_list, axis=0)
+            rhand_bbox_valid = np.stack(rhand_bbox_valid_list, axis=0)
+            rhand_bbox_size = np.stack(rhand_bbox_size_list, axis=0)
+
+
+            if 'occlusion' in data:
+                occlusion = data['occlusion']
+                occ_mask = occlusion<97
+                
+                joint_img_aug[:,:,2] = joint_img_aug[:,:,2]*occ_mask[:,None]
+                joint_cam_wo_ra[:,:,3] = joint_cam_wo_ra[:,:,3]*occ_mask[:,None]
+                joint_trunc = joint_trunc*occ_mask[:,None,None]
+                smplx_pose_valid = smplx_pose_valid*occ_mask[:,None]
+                smplx_joint_valid = smplx_joint_valid*occ_mask[:,None,None]
+                smplx_expr_valid = smplx_expr_valid*occ_mask
+                smplx_shape_valid = smplx_shape_valid*occ_mask
+                rhand_bbox_valid = rhand_bbox_valid*occ_mask
+                lhand_bbox_valid = lhand_bbox_valid*occ_mask
+                face_bbox_valid = face_bbox_valid*occ_mask
+                
+            
+            if 'is_kid' in data:
+                is_kid = data['is_kid'].copy()
+                smplx_shape_valid = smplx_shape_valid * (is_kid==0)
+                
+                
+            inputs = {'img': img}
+
+            joint_img_aug[:,:,2] = joint_img_aug[:,:,2] * body_bbox_valid[:,None]
+            
+            is_3D = float(False) if dummy_cord else float(True)
+            
+            targets = {
+                # keypoints2d, [0,img_w],[0,img_h] -> [0,1] -> [0,output_hm_shape]
+                'joint_img': joint_img_aug[body_bbox_valid>0], 
+                # joint_cam, kp3d wo ra # raw kps3d probably without ra
+                'joint_cam': joint_cam_wo_ra[body_bbox_valid>0], 
+                # kps3d with body, face, hand ra
+                'smplx_joint_cam': joint_cam_ra[body_bbox_valid>0], 
+                'smplx_pose': smplx_pose[body_bbox_valid>0],
+                'smplx_shape': smplx_shape[body_bbox_valid>0],
+                'smplx_expr': smplx_expr[body_bbox_valid>0],
+                'lhand_bbox_center': lhand_bbox_center[body_bbox_valid>0], 
+                'lhand_bbox_size': lhand_bbox_size[body_bbox_valid>0],
+                'rhand_bbox_center': rhand_bbox_center[body_bbox_valid>0], 
+                'rhand_bbox_size': rhand_bbox_size[body_bbox_valid>0],
+                'face_bbox_center': face_bbox_center[body_bbox_valid>0], 
+                'face_bbox_size': face_bbox_size[body_bbox_valid>0],
+                'body_bbox_center': body_bbox_center[body_bbox_valid>0], 
+                'body_bbox_size': body_bbox_size[body_bbox_valid>0],
+                'body_bbox': body_bbox.reshape(-1,4)[body_bbox_valid>0],
+                'lhand_bbox': lhand_bbox.reshape(-1,4)[body_bbox_valid>0],
+                'rhand_bbox': rhand_bbox.reshape(-1,4)[body_bbox_valid>0],
+                'face_bbox': face_bbox.reshape(-1,4)[body_bbox_valid>0],
+                'gender': gender[body_bbox_valid>0]}
+            
+            meta_info = {
+                'joint_trunc': joint_trunc[body_bbox_valid>0],
+                'smplx_pose_valid': smplx_pose_valid[body_bbox_valid>0],
+                'smplx_shape_valid': smplx_shape_valid[body_bbox_valid>0],
+                'smplx_expr_valid': smplx_expr_valid[body_bbox_valid>0],
+                'is_3D': is_3D, 
+                'lhand_bbox_valid': lhand_bbox_valid[body_bbox_valid>0],
+                'rhand_bbox_valid': rhand_bbox_valid[body_bbox_valid>0], 
+                'face_bbox_valid': face_bbox_valid[body_bbox_valid>0],
+                'body_bbox_valid': body_bbox_valid[body_bbox_valid>0],
+                'img_shape': np.array(img.shape[:2]), 
+                'ori_shape':data['img_shape'],
+                'idx': idx
+               
+            }            
+            result = {**inputs, **targets, **meta_info}
+            
+            result = self.normalize(result)
+            result = self.format(result)
+            return result
+
+        
+
+        if self.data_split == 'test':
+            self.cam_param = {}
+            joint_cam = data['joint_cam']
+            
+            if joint_cam is not None:
+                dummy_cord = False
+                joint_cam[:,:,:3] = joint_cam[:,:,:3] - joint_cam[
+                    :, self.joint_set['root_joint_idx'], None, :3]  # root-relative
+            else:
+                # dummy cord as joint_cam
+                dummy_cord = True
+                joint_cam = np.zeros(
+                    (num_person, self.joint_set['joint_num'], 3),
+                                     dtype=np.float32)
+
+            joint_img = data['joint_img']
+            
+            
+            joint_img_aug, joint_cam_wo_ra, joint_cam_ra, joint_trunc = \
+                process_db_coord_batch_no_valid(
+                    joint_img, joint_cam, do_flip, img_shape,
+                    self.joint_set['flip_pairs'], img2bb_trans, rot,
+                    self.joint_set['joints_name'], smpl_x.joints_name,
+                    cropped_img_shape)
+            
+            
+
+            # smplx coordinates and parameters
+            smplx_param = data['smplx_param']
+            # smplx_cam_trans = np.array(
+            #     smplx_param['trans']) if 'trans' in smplx_param else None
+            # TODO: remove this, seperate smpl and smplx
+            smplx_pose, smplx_shape, smplx_expr, smplx_pose_valid, \
+            smplx_joint_valid, smplx_expr_valid, smplx_shape_valid = \
+                process_human_model_output_batch_simplify(
+                    smplx_param, do_flip, rot, as_smplx)
+            
+            # if cam not provided, we take joint_img as smplx joint 2d, 
+            # which is commonly the case for our processed humandata
+            if self.use_betas_neutral:
+                smplx_shape = smplx_param['betas_neutral'].reshape(
+                    num_person, -1)
+                smplx_shape[(np.abs(smplx_shape) > 3).any(axis=1)] = 0.
+                smplx_shape = smplx_shape.reshape(num_person, -1)
+            
+            smplx_joint_valid = smplx_joint_valid[:, :, None]
+            
+            lhand_bbox_center_list = []
+            lhand_bbox_valid_list = []
+            lhand_bbox_size_list = []
+            lhand_bbox_list = []
+            face_bbox_center_list = []
+            face_bbox_size_list = []
+            face_bbox_valid_list = []
+            face_bbox_list = []
+            rhand_bbox_center_list = []
+            rhand_bbox_valid_list = []
+            rhand_bbox_size_list = []
+            rhand_bbox_list = []
+            body_bbox_center_list = []
+            body_bbox_size_list = []
+            body_bbox_valid_list = []
+            body_bbox_list = []
+                        
+            for i in range(num_person):
+                lhand_bbox, lhand_bbox_valid = self.process_hand_face_bbox(
+                    data['lhand_bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                rhand_bbox, rhand_bbox_valid = self.process_hand_face_bbox(
+                    data['rhand_bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                face_bbox, face_bbox_valid = self.process_hand_face_bbox(
+                    data['face_bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)
+                
+                body_bbox, body_bbox_valid = self.process_hand_face_bbox(
+                    data['bbox'][i], do_flip, img_shape, img2bb_trans,
+                    cropped_img_shape)                
+
+                if do_flip:
+                    lhand_bbox, rhand_bbox = rhand_bbox, lhand_bbox
+                    lhand_bbox_valid, rhand_bbox_valid = rhand_bbox_valid, lhand_bbox_valid            
+
+                body_bbox_list.append(body_bbox)
+                lhand_bbox_list.append(lhand_bbox)
+                rhand_bbox_list.append(rhand_bbox)
+                face_bbox_list.append(face_bbox)
+
+                lhand_bbox_center = (lhand_bbox[0] + lhand_bbox[1]) / 2.
+                rhand_bbox_center = (rhand_bbox[0] + rhand_bbox[1]) / 2.
+                face_bbox_center = (face_bbox[0] + face_bbox[1]) / 2.
+                body_bbox_center = (body_bbox[0] + body_bbox[1]) / 2.
+                lhand_bbox_size = lhand_bbox[1] - lhand_bbox[0]
+                rhand_bbox_size = rhand_bbox[1] - rhand_bbox[0]
+
+                face_bbox_size = face_bbox[1] - face_bbox[0]
+                body_bbox_size = body_bbox[1] - body_bbox[0]
+                lhand_bbox_center_list.append(lhand_bbox_center)
+                lhand_bbox_valid_list.append(lhand_bbox_valid)
+                lhand_bbox_size_list.append(lhand_bbox_size)
+                face_bbox_center_list.append(face_bbox_center)
+                face_bbox_size_list.append(face_bbox_size)
+                face_bbox_valid_list.append(face_bbox_valid)
+                rhand_bbox_center_list.append(rhand_bbox_center)
+                rhand_bbox_valid_list.append(rhand_bbox_valid)
+                rhand_bbox_size_list.append(rhand_bbox_size)
+                body_bbox_center_list.append(body_bbox_center)
+                body_bbox_size_list.append(body_bbox_size)
+                body_bbox_valid_list.append(body_bbox_valid)
+
+            body_bbox = np.stack(body_bbox_list, axis=0)
+            lhand_bbox = np.stack(lhand_bbox_list, axis=0)
+            rhand_bbox = np.stack(rhand_bbox_list, axis=0)
+            face_bbox = np.stack(face_bbox_list, axis=0)
+            lhand_bbox_center = np.stack(lhand_bbox_center_list, axis=0)
+            lhand_bbox_valid = np.stack(lhand_bbox_valid_list, axis=0)
+            lhand_bbox_size = np.stack(lhand_bbox_size_list, axis=0)
+            face_bbox_center = np.stack(face_bbox_center_list, axis=0)
+            face_bbox_size = np.stack(face_bbox_size_list, axis=0)
+            face_bbox_valid = np.stack(face_bbox_valid_list, axis=0)
+            body_bbox_center = np.stack(body_bbox_center_list, axis=0)
+            body_bbox_size = np.stack(body_bbox_size_list, axis=0)
+            body_bbox_valid = np.stack(body_bbox_valid_list, axis=0)
+            rhand_bbox_center = np.stack(rhand_bbox_center_list, axis=0)
+            rhand_bbox_valid = np.stack(rhand_bbox_valid_list, axis=0)
+            rhand_bbox_size = np.stack(rhand_bbox_size_list, axis=0)
+                                            
+                            
+            inputs = {'img': img}
+            
+            targets = {
+                # keypoints2d, [0,img_w],[0,img_h] -> [0,1] -> [0,output_hm_shape]
+                'joint_img': joint_img_aug, 
+                # projected smplx if valid cam_param, else same as keypoints2d
+                # joint_cam, kp3d wo ra # raw kps3d probably without ra
+                'joint_cam': joint_cam_wo_ra, 
+                'ann_idx': idx,
+                # kps3d with body, face, hand ra
+                'smplx_joint_cam': joint_cam_ra,
+                'smplx_pose': smplx_pose,
+                'smplx_shape': smplx_shape,
+                'smplx_expr': smplx_expr,
+                'lhand_bbox_center': lhand_bbox_center, 
+                'lhand_bbox_size': lhand_bbox_size,
+                'rhand_bbox_center': rhand_bbox_center, 
+                'rhand_bbox_size': rhand_bbox_size,
+                'face_bbox_center': face_bbox_center, 
+                'face_bbox_size': face_bbox_size,
+                'body_bbox_center': body_bbox_center, 
+                'body_bbox_size': body_bbox_size,
+                'body_bbox': body_bbox.reshape(-1,4),
+                'lhand_bbox': lhand_bbox.reshape(-1,4),
+                'rhand_bbox': rhand_bbox.reshape(-1,4),
+                'face_bbox': face_bbox.reshape(-1,4),
+                'gender': gender,
+                'bb2img_trans': bb2img_trans,
+            }
+            
+            if self.body_only:
+                meta_info = {
+                    'joint_trunc': joint_trunc,
+                    'smplx_pose_valid': smplx_pose_valid,
+                    'smplx_shape_valid': float(smplx_shape_valid),
+                    'smplx_expr_valid': smplx_expr_valid,
+                    'is_3D': float(False) if dummy_cord else float(True), 
+                    'lhand_bbox_valid': lhand_bbox_valid,
+                    'rhand_bbox_valid': rhand_bbox_valid, 
+                    'face_bbox_valid': face_bbox_valid,
+                    'body_bbox_valid': body_bbox_valid,
+                    'img_shape': np.array(img.shape[:2]), 
+                    'ori_shape':data['img_shape'],
+                    'idx': idx
+                }
+            else:
+                meta_info = {
+                    'joint_trunc': joint_trunc,
+                    'smplx_pose_valid': smplx_pose_valid,
+                    'smplx_shape_valid': smplx_shape_valid,
+                    'smplx_expr_valid': smplx_expr_valid,
+                    'is_3D': float(False) if dummy_cord else float(True), 
+                    'lhand_bbox_valid': lhand_bbox_valid,
+                    'rhand_bbox_valid': rhand_bbox_valid, 
+                    'face_bbox_valid': face_bbox_valid,
+                    'body_bbox_valid': body_bbox_valid,
+                    'img_shape': np.array(img.shape[:2]), 
+                    'ori_shape':data['img_shape'],
+                    'idx': idx
+                   }
+            
+            result = {**inputs, **targets, **meta_info}
+            result = self.normalize(result)
+            result = self.format(result)
+            return result
+        
+    def evaluate(self, outs, cur_sample_idx):
+        annots = self.datalist
+        sample_num = len(outs)
+        eval_result = {
+            'pa_mpvpe_all': [],
+            'pa_mpvpe_l_hand': [],
+            'pa_mpvpe_r_hand': [],
+            'pa_mpvpe_hand': [],
+            'pa_mpvpe_face': [],
+            'mpvpe_all': [],
+            'mpvpe_l_hand': [],
+            'mpvpe_r_hand': [],
+            'mpvpe_hand': [],
+            'mpvpe_face': []
+        }
+
+        vis = getattr(cfg, 'vis', False)
+        vis_save_dir = cfg.vis_dir
+        
+        csv_file = f'{cfg.result_dir}/agora_smplx_error.csv'
+        file = open(csv_file, 'a', newline='')
+        for n in range(sample_num):
+            annot = annots[cur_sample_idx + n]
+            out = outs[n]
+            mesh_gt = out['smplx_mesh_cam_target']
+            mesh_out = out['smplx_mesh_cam']
+            
+            # print('zzz',mesh_gt.shape,mesh_out.shape)
+            # from pytorch3d.io import save_obj
+            # for m_i,(mesh_gt_i,mesh_out_i) in enumerate(zip(mesh_gt,mesh_out)):
+            #     save_obj('temp_gt_%d.obj'%m_i,verts=torch.Tensor(mesh_gt_i),faces=torch.tensor([]))
+            #     save_obj('temp_pred_%d.obj'%m_i,verts=torch.Tensor(mesh_out_i),faces=torch.tensor([]))
+            
+            ann_idx = out['gt_ann_idx']
+            img_path = []
+            for ann_id in ann_idx:
+                img_path.append(annots[ann_id]['img_path'])
+            eval_result['img_path'] = img_path
+            eval_result['ann_idx'] = ann_idx
+            # MPVPE from all vertices
+            mesh_out_align = \
+                mesh_out - np.dot(
+                    smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['pelvis'], None, :] + \
+                    np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['pelvis'], None, :]
+            
+            eval_result['mpvpe_all'].extend(
+                np.sqrt(np.sum(
+                    (mesh_out_align - mesh_gt)**2, -1)).mean(-1) * 1000)
+            mesh_out_align = rigid_align_batch(mesh_out, mesh_gt)
+            eval_result['pa_mpvpe_all'].extend(
+                np.sqrt(np.sum(
+                    (mesh_out_align - mesh_gt)**2, -1)).mean(-1) * 1000)
+
+            # MPVPE from hand vertices
+            mesh_gt_lhand = mesh_gt[:, smpl_x.hand_vertex_idx['left_hand'], :]
+            mesh_out_lhand = mesh_out[:, smpl_x.hand_vertex_idx['left_hand'], :]
+            mesh_gt_rhand = mesh_gt[:, smpl_x.hand_vertex_idx['right_hand'], :]
+            mesh_out_rhand = mesh_out[:, smpl_x.hand_vertex_idx['right_hand'], :]
+            mesh_out_lhand_align = \
+                mesh_out_lhand - \
+                np.dot(smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['lwrist'], None, :] + \
+                np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['lwrist'], None, :]
+                    
+            mesh_out_rhand_align = \
+                mesh_out_rhand - \
+                np.dot(smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['rwrist'], None, :] + \
+                np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['rwrist'], None, :]
+            
+            eval_result['mpvpe_l_hand'].extend(
+                np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean(-1) *
+                1000)
+            eval_result['mpvpe_r_hand'].extend(
+                np.sqrt(np.sum(
+                    (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean(-1) *
+                1000)
+            eval_result['mpvpe_hand'].extend(
+                (np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean(-1) *
+                 1000 +
+                 np.sqrt(np.sum(
+                     (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean(-1) *
+                 1000) / 2.)
+            mesh_out_lhand_align = rigid_align_batch(mesh_out_lhand, mesh_gt_lhand)
+            mesh_out_rhand_align = rigid_align_batch(mesh_out_rhand, mesh_gt_rhand)
+            eval_result['pa_mpvpe_l_hand'].extend(
+                np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean(-1) *
+                1000)
+            eval_result['pa_mpvpe_r_hand'].extend(
+                np.sqrt(np.sum(
+                    (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean(-1) *
+                1000)
+            eval_result['pa_mpvpe_hand'].extend(
+                (np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean(-1) *
+                 1000 +
+                 np.sqrt(np.sum(
+                     (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean(-1) *
+                 1000) / 2.)
+            
+            
+            save_error=True
+            if save_error:
+                writer = csv.writer(file)
+                new_line = [ann_idx[n],img_path[n], eval_result['mpvpe_all'][-1], eval_result['pa_mpvpe_all'][-1]]
+                writer.writerow(new_line)
+                self.save_idx += 1
+            
+            
+        return eval_result
+
+
+    def print_eval_result(self, eval_result):
+
+        print('AGORA test results are dumped at: ' +
+              osp.join(cfg.result_dir, 'predictions'))
+
+        if self.data_split == 'test' and self.test_set == 'test':  # do not print. just submit the results to the official evaluation server
+            return
+
+        print('======AGORA-val======')
+        print('PA MPVPE (All): %.2f mm' % np.mean(eval_result['pa_mpvpe_all']))
+        print('PA MPVPE (L-Hands): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_l_hand']))
+        print('PA MPVPE (R-Hands): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_r_hand']))
+        print('PA MPVPE (Hands): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_hand']))
+        print('PA MPVPE (Face): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_face']))
+        print()
+
+        print('MPVPE (All): %.2f mm' % np.mean(eval_result['mpvpe_all']))
+        print('MPVPE (L-Hands): %.2f mm' %
+              np.mean(eval_result['mpvpe_l_hand']))
+        print('MPVPE (R-Hands): %.2f mm' %
+              np.mean(eval_result['mpvpe_r_hand']))
+        print('MPVPE (Hands): %.2f mm' % np.mean(eval_result['mpvpe_hand']))
+        print('MPVPE (Face): %.2f mm' % np.mean(eval_result['mpvpe_face']))
+        
+        out_file = osp.join(cfg.result_dir,'agora_val.txt')
+        if os.path.exists(out_file):
+            f = open(out_file, 'a+')
+        else:
+            f = open(out_file, 'w', encoding="utf-8")
+            
+        f.write('\n')
+        f.write(f'{cfg.exp_name}\n')            
+        f.write(f'AGORA-val dataset: \n')
+        f.write('PA MPVPE (All): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_all']))
+        f.write('PA MPVPE (L-Hands): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_l_hand']))
+        f.write('PA MPVPE (R-Hands): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_r_hand']))
+        f.write('PA MPVPE (Hands): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_hand']))
+        f.write('PA MPVPE (Face): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_face']))
+        f.write('MPVPE (All): %.2f mm\n' % np.mean(eval_result['mpvpe_all']))
+        f.write('MPVPE (L-Hands): %.2f mm\n' %
+                np.mean(eval_result['mpvpe_l_hand']))
+        f.write('MPVPE (R-Hands): %.2f mm\n' %
+                np.mean(eval_result['mpvpe_r_hand']))
+        f.write('MPVPE (Hands): %.2f mm\n' % np.mean(eval_result['mpvpe_hand']))
+        f.write('MPVPE (Face): %.2f mm\n' % np.mean(eval_result['mpvpe_face']))

datasets/ARCTIC.py

@@ -0,0 +1,215 @@
+import os
+import os.path as osp
+from glob import glob
+import numpy as np
+from config.config import cfg
+
+import csv
+
+from util.human_models import smpl_x
+
+from util.transforms import rigid_align_batch
+
+from humandata import HumanDataset
+
+class ARCTIC(HumanDataset):
+    def __init__(self, transform, data_split):
+        super(ARCTIC, self).__init__(transform, data_split)
+
+        self.img_dir = 'data/osx_data/ARCTIC'
+
+
+        if data_split == 'train':
+            self.annot_path = 'data/preprocessed_npz/multihuman_data/p1_train_multi.npz'
+            self.annot_path_cache = 'data/preprocessed_npz/cache/p1_train_cache_sample1000_080824.npz'
+            self.sample_interval = 1000  
+        elif data_split == 'test':
+            self.annot_path = 'data/preprocessed_npz_old/multihuman_data/p1_val_multi.npz'
+            self.annot_path_cache = 'data/preprocessed_npz_old/cache/p1_val_cache_30.npz'
+            self.sample_interval = 30
+        
+        
+        self.use_cache = getattr(cfg, 'use_cache', False)
+        self.img_shape = None  #1024, 1024)  # (h, w)
+        self.cam_param = {}
+        self.use_cache=True
+        # load data
+        if self.use_cache and osp.isfile(self.annot_path_cache):
+            print(
+                f'[{self.__class__.__name__}] loading cache from {self.annot_path_cache}'
+            )
+            self.datalist = self.load_cache(self.annot_path_cache)
+        else:
+            if self.use_cache:
+                print(
+                    f'[{self.__class__.__name__}] Cache not found, generating cache...'
+                )
+            self.datalist = self.load_data(train_sample_interval=getattr(
+                cfg, f'{self.__class__.__name__}_train_sample_interval', self.sample_interval))
+            if self.use_cache:
+                self.save_cache(self.annot_path_cache, self.datalist)
+                
+
+    def evaluate(self, outs, cur_sample_idx):
+        annots = self.datalist
+        sample_num = len(outs)
+        eval_result = {
+            'pa_mpvpe_all': [],
+            'pa_mpvpe_l_hand': [],
+            'pa_mpvpe_r_hand': [],
+            'pa_mpvpe_hand': [],
+            'pa_mpvpe_face': [],
+            'mpvpe_all': [],
+            'mpvpe_l_hand': [],
+            'mpvpe_r_hand': [],
+            'mpvpe_hand': [],
+            'mpvpe_face': []
+        }
+
+        vis = getattr(cfg, 'vis', False)
+        vis_save_dir = cfg.vis_dir
+        csv_file = f'{cfg.result_dir}/arctic_smplx_error.csv'
+        file = open(csv_file, 'a', newline='')
+        
+        for n in range(sample_num):
+            annot = annots[cur_sample_idx + n]
+            out = outs[n]
+            mesh_gt = out['smplx_mesh_cam_target']
+            mesh_out = out['smplx_mesh_cam']
+            ann_idx = out['gt_ann_idx']
+            img_path = []
+            for ann_id in ann_idx:
+                img_path.append(annots[ann_id]['img_path'])
+            eval_result['img_path'] = img_path
+            # MPVPE from all vertices
+            mesh_out_align = \
+                mesh_out - np.dot(
+                    smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['pelvis'], None, :] + \
+                    np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['pelvis'], None, :]
+            
+            eval_result['mpvpe_all'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_align - mesh_gt)**2, -1)).mean() * 1000)
+            mesh_out_align = rigid_align_batch(mesh_out, mesh_gt)
+            eval_result['pa_mpvpe_all'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_align - mesh_gt)**2, -1)).mean() * 1000)
+
+            # MPVPE from hand vertices
+            mesh_gt_lhand = mesh_gt[:, smpl_x.hand_vertex_idx['left_hand'], :]
+            mesh_out_lhand = mesh_out[:, smpl_x.hand_vertex_idx['left_hand'], :]
+            mesh_gt_rhand = mesh_gt[:, smpl_x.hand_vertex_idx['right_hand'], :]
+            mesh_out_rhand = mesh_out[:, smpl_x.hand_vertex_idx['right_hand'], :]
+            mesh_out_lhand_align = \
+                mesh_out_lhand - \
+                np.dot(smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['lwrist'], None, :] + \
+                np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['lwrist'], None, :]
+                    
+            mesh_out_rhand_align = \
+                mesh_out_rhand - \
+                np.dot(smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['rwrist'], None, :] + \
+                np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['rwrist'], None, :]
+            
+            eval_result['mpvpe_l_hand'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean() *
+                1000)
+            eval_result['mpvpe_r_hand'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean() *
+                1000)
+            eval_result['mpvpe_hand'].append(
+                (np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean() *
+                 1000 +
+                 np.sqrt(np.sum(
+                     (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean() *
+                 1000) / 2.)
+            mesh_out_lhand_align = rigid_align_batch(mesh_out_lhand, mesh_gt_lhand)
+            mesh_out_rhand_align = rigid_align_batch(mesh_out_rhand, mesh_gt_rhand)
+            eval_result['pa_mpvpe_l_hand'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean() *
+                1000)
+            eval_result['pa_mpvpe_r_hand'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean() *
+                1000)
+            eval_result['pa_mpvpe_hand'].append(
+                (np.sqrt(np.sum(
+                    (mesh_out_lhand_align - mesh_gt_lhand)**2, -1)).mean() *
+                 1000 +
+                 np.sqrt(np.sum(
+                     (mesh_out_rhand_align - mesh_gt_rhand)**2, -1)).mean() *
+                 1000) / 2.)
+
+            # MPVPE from face vertices
+            mesh_gt_face = mesh_gt[:, smpl_x.face_vertex_idx, :]
+            mesh_out_face = mesh_out[:, smpl_x.face_vertex_idx, :]
+            mesh_out_face_align = \
+                mesh_out_face - \
+                np.dot(smpl_x.J_regressor, mesh_out).transpose(1,0,2)[:, smpl_x.J_regressor_idx['neck'], None, :] + \
+                np.dot(smpl_x.J_regressor, mesh_gt).transpose(1,0,2)[:, smpl_x.J_regressor_idx['neck'], None, :]
+            eval_result['mpvpe_face'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_face_align - mesh_gt_face)**2, -1)).mean() * 1000)
+            mesh_out_face_align = rigid_align_batch(mesh_out_face, mesh_gt_face)
+            eval_result['pa_mpvpe_face'].append(
+                np.sqrt(np.sum(
+                    (mesh_out_face_align - mesh_gt_face)**2, -1)).mean() * 1000)
+        
+            save_error=True
+            if save_error:
+                writer = csv.writer(file)
+                new_line = [ann_idx[n], img_path[n], eval_result['mpvpe_all'][-1], eval_result['pa_mpvpe_all'][-1]]
+                writer.writerow(new_line)
+                # self.save_idx += 1
+        return eval_result
+
+    def print_eval_result(self, eval_result):
+
+        print('======ARCTIC-val======')
+        print('PA MPVPE (All): %.2f mm' % np.mean(eval_result['pa_mpvpe_all']))
+        print('PA MPVPE (L-Hands): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_l_hand']))
+        print('PA MPVPE (R-Hands): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_r_hand']))
+        print('PA MPVPE (Hands): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_hand']))
+        print('PA MPVPE (Face): %.2f mm' %
+              np.mean(eval_result['pa_mpvpe_face']))
+        print()
+
+        print('MPVPE (All): %.2f mm' % np.mean(eval_result['mpvpe_all']))
+        print('MPVPE (L-Hands): %.2f mm' %
+              np.mean(eval_result['mpvpe_l_hand']))
+        print('MPVPE (R-Hands): %.2f mm' %
+              np.mean(eval_result['mpvpe_r_hand']))
+        print('MPVPE (Hands): %.2f mm' % np.mean(eval_result['mpvpe_hand']))
+        print('MPVPE (Face): %.2f mm' % np.mean(eval_result['mpvpe_face']))
+
+        out_file = osp.join(cfg.result_dir,'arctic_val.txt')
+        if os.path.exists(out_file):
+            f = open(out_file, 'a+')
+        else:
+            f = open(out_file, 'w', encoding="utf-8")
+        f.write('\n')
+        f.write(f'{cfg.exp_name}\n')   
+        f.write(f'ARCTIC-val dataset: \n')
+        f.write('PA MPVPE (All): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_all']))
+        f.write('PA MPVPE (L-Hands): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_l_hand']))
+        f.write('PA MPVPE (R-Hands): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_r_hand']))
+        f.write('PA MPVPE (Hands): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_hand']))
+        f.write('PA MPVPE (Face): %.2f mm\n' %
+                np.mean(eval_result['pa_mpvpe_face']))
+        f.write('MPVPE (All): %.2f mm\n' % np.mean(eval_result['mpvpe_all']))
+        f.write('MPVPE (L-Hands): %.2f mm\n' %
+                np.mean(eval_result['mpvpe_l_hand']))
+        f.write('MPVPE (R-Hands): %.2f mm\n' %
+                np.mean(eval_result['mpvpe_r_hand']))
+        f.write('MPVPE (Hands): %.2f mm\n' % np.mean(eval_result['mpvpe_hand']))
+        f.write('MPVPE (Face): %.2f mm\n' % np.mean(eval_result['mpvpe_face']))

datasets/BEDLAM.py

@@ -0,0 +1,32 @@
+import os.path as osp
+from config.config import cfg
+from humandata import HumanDataset
+
+
+class BEDLAM(HumanDataset):
+    def __init__(self, transform, data_split):
+        super(BEDLAM, self).__init__(transform, data_split)
+
+        self.img_dir = './data/datasets/bedlam/train_images/'
+        self.annot_path = 'data/preprocessed_npz/multihuman_data/bedlam_train_multi_0915.npz'
+        self.annot_path_cache = 'data/preprocessed_npz/cache/bedlam_train_cache_080824.npz'        
+        self.use_cache = getattr(cfg, 'use_cache', False)
+        
+        self.img_shape = None  #1024, 1024)  # (h, w)
+        self.cam_param = {}
+
+        # load data or cache
+        if self.use_cache and osp.isfile(self.annot_path_cache):
+            print(
+                f'[{self.__class__.__name__}] loading cache from {self.annot_path_cache}'
+            )
+            self.datalist = self.load_cache(self.annot_path_cache)
+        else:
+            if self.use_cache:
+                print(
+                    f'[{self.__class__.__name__}] Cache not found, generating cache...'
+                )
+            self.datalist = self.load_data(train_sample_interval=getattr(
+                cfg, f'{self.__class__.__name__}_train_sample_interval', 5))
+            if self.use_cache:
+                self.save_cache(self.annot_path_cache, self.datalist)