-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathtemp_readme.txt
More file actions
92 lines (45 loc) · 4.65 KB
/
temp_readme.txt
File metadata and controls
92 lines (45 loc) · 4.65 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
ros2 launch roverrobotics_gazebo Leo_rover_gazebo.launch.py
## ros2 launch roverrobotics_gazebo 4wd_rover_gazebo.launch.py
~/src/RoboTerrain/ros2_ws/src$ python ign_ros2_pose_topic.py inspect leo_rover
~/src/RoboTerrain/ros2_ws/src$ python ign_ros2_pose_topic.py inspect rover_zero4wd
~/src/attention/inference$ python ros2_mem_share.py
/dynamic_obstacles$ python spawn.py trajectory_file_name actor_name walk_name world_name
python spawn.py --trajectory_file inspect_corner_triangle.sdf --world_name inspect
python inference.py --attention_mode ./checkpoints/checkpoint_epoch_
############## Dreamerv3 commands
FILTERED_LD_LIBRARY_PATH=$(echo $LD_LIBRARY_PATH | tr ':' '\n' | grep -E '^/opt/ros' | tr '\n' ':' | sed 's/:$//')
env LD_LIBRARY_PATH="$FILTERED_LD_LIBRARY_PATH" CUDA_HOME="" XLA_PYTHON_CLIENT_PREALLOCATE=false XLA_PYTHON_CLIENT_ALLOCATOR=platform python dreamerv3/main.py --configs leorover --logdir ~/backup_500GB/logdir/dreamer/{timestamp}
python run_efficient_training.py --dataset_path /home/jack/data/social_nav --num_epochs 1000 --batch_size 4 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path models/yolo11n.onnx --embedding_dim 128 --num_heads 4 --resume_checkpoint checkpoints/checkpoint_epoch
python run_efficient_training.py --dataset_path /home/jack/data/social_nav --num_epochs 1000 --batch_size 4 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path models/yolo11n.onnx --embedding_dim 128 --num_heads 4 --resume_checkpoint checkpoints/checkpoint_epoch_2351.pkl
(sb3) jack@HAL:~/src/RoboTerrain/ros2_ws/src/sb3$ python sb3_SAC.py --mode train --load False --world inspect --vision True
ros2 run teleop_twist_keyboard teleop_twist_keyboard
ros2 run rqt_image_view rqt_image_view
###################### fixes!!:
-change publisher to match frame rate of sim not wall clock time.
-negative reward for pointing toward human or prediction
-continuing training of attention mechanism
-test LeSTA
# Show sizes of all directories in current folder, sorted:
du -h --max-depth=1 | sort -h
du -sh ./*
camera test
ros2 launch roverrobotics_gazebo 4wd_rover_gazebo.launch.py
python run_efficient_training.py --dataset_path /home/jack/data/social_nav --num_epochs 1000 --batch_size 4 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path ../models/yolo11n.onnx --embedding_dim 128 --num_heads 4
python run_efficient_training.py --dataset_path /home/jack/data/social_nav --num_epochs 1000 --batch_size 4 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path ../models/yolo11n.onnx --embedding_dim 128 --num_heads 4 --resume_checkpoint ./model_output/checkpoint_epoch_1501.pkl
python cleanup_shm.py --all --force
python run_efficient_training.py --dataset_path /home/jack/data/social_nav/crossroad --num_epochs 30
python run_efficient_training.py --dataset_path /home/jack/data/social_nav/crossroad --num_epochs 30 --batch_size 8 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path /path/to/your/yolo11n.onnx --embedding_dim 128 --num_heads 4
python run_efficient_training.py --dataset_path /home/jack/data/social_nav/crossroad --num_epochs 30
python run_trajectory_prediction.py --mode train --dataset_path /home/jack/data/social_nav/alley/ --num_epochs 1 --target_width 320 --target_height 320
unset LD_LIBRARY_PATH
export XLA_PYTHON_CLIENT_PREALLOCATE=false
export XLA_PYTHON_CLIENT_MEM_FRACTION=.30
python run_efficient_training.py --dataset_path /home/jack/data/social_nav/subway --num_epochs 30 --batch_size 8 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path ../models/yolo11n.onnx --embedding_dim 128 --num_heads 4
(attent) jack@HAL:~/src/attention$ python run_efficient_training.py --dataset_path /home/jack/data/social_nav --num_epochs 1000 --batch_size 1 --sequence_length 5 --target_width 320 --target_height 320 --yolo_model_path ../models/yolo11n.onnx --embedding_dim 128 --num_heads 4 --resume_checkpoint checkpoints/checkpoint_epoch_2351.pkl
xFormers not available
xFormers not available
2025-06-22 14:16:46,375 - root - INFO - Preprocessing dataset at /home/jack/data/social_nav using memory mapping
2025-06-22 14:16:46,375 - root - INFO - Preprocessed dataset already exists at /home/jack/src/attention/preprocessed_data/social_nav_seq5_stride1_w320_h320_max1000.npz
2025-06-22 14:16:46,375 - root - INFO - Preprocessing dataset at /home/jack/data/social_nav using memory mapping
2025-06-22 14:16:46,375 - root - INFO - Preprocessed dataset already exists at /home/jack/src/attention/preprocessed_data/social_nav_seq5_stride2_w320_h320_max500.npz
2025-06-22 14:16:46,375 - root - INFO - Starting efficient training with 1000 epochs