-
Notifications
You must be signed in to change notification settings - Fork 0
/
visualize_real.py
277 lines (245 loc) · 11.4 KB
/
visualize_real.py
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
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
import mj_envs
import click
import os
import gym
import json
import torch
import numpy as np
import pickle
import time
import matplotlib.pyplot as plt
import torchvision.transforms as T
from einops import rearrange
from PIL import Image
# from mjrl.utils.gym_env import GymEnv
# from src.models.imi_models_real import Actor_share,Actor,Actor_share_score,Actor_share_new
from src.models.SoftGrasp_models_real import Actor
from src.models.encoders import (
make_image_encoder,
make_torque_encoder,
make_angle_encoder,
make_share_encoder,
make_share_POS_encoder,
make_torque_Proprioceptionencoder,
make_angle_Proprioceptionencoder,
)
from torch.utils.data import DataLoader,RandomSampler,IterableDataset
from hand_relocate_v1_mujoco import make_sim_env
from PIL import Image
# from robot.real_env import make_real_env_hand_arm,make_real_env_hand
from robot.real_env import make_real_env_hand
import cv2
# from matplotlib import rcParams
# config = {
# "font.family":'serif',
# "font.serif":['Times'],
# "font.size": 14,
# "mathtext.fontset":'stix',
# "font.serif": ['SimSun'],
# 'text.usetex':True
# }
# rcParams.update(config)
def main(args):
camera_names = args.camera_names
resized_height_v = args.resized_height_v
resized_width_v = args.resized_width_v
resized_height_t = args.resized_height_t
resized_width_t = args.resized_width_t
I_encoder = make_image_encoder(args.encoder_dim)
if "mlp" in args.encoder:
if args.use_one_hot:
T_encoder = make_torque_Proprioceptionencoder(args.one_hot_torque_dim, args.encoder_dim )
A_encoder = make_angle_Proprioceptionencoder(args.one_hot_angle_dim, args.encoder_dim )
share_encoder = make_share_POS_encoder(args.share_dim, args.encoder_dim )
else:
T_encoder = make_torque_Proprioceptionencoder(args.torque_dim, args.encoder_dim )
A_encoder = make_angle_Proprioceptionencoder(args.angle_dim, args.encoder_dim )
share_encoder = make_share_POS_encoder(args.share_dim, args.encoder_dim )
else:
T_encoder = make_torque_encoder(args.torque_dim, args.encoder_dim * args.num_stack )
A_encoder = make_angle_encoder(args.angle_dim, args.encoder_dim * args.num_stack )
share_encoder = make_share_encoder(args.torque_dim , args.encoder_dim * args.num_stack)
modalities = args.ablation.split("_")
actor = Actor(I_encoder,T_encoder,A_encoder,share_encoder, args).cuda()
if args.real_robot:
e = make_real_env_hand(init_node=True, setup_robots=True, setup_base=True)
else:
e = make_sim_env()
# 加载ckpt文件中的模型参数
modalities = args.ablation.split("_")
print('modalities',modalities)
print('len(modalities) ',len(modalities) )
if len(modalities) == 3 :
if "a_mha" in args.use_way:
if args.use_pos:
checkpoint = torch.load(args.j_f_vf_m_pos_ckpt)
print(1111)
else :
checkpoint = torch.load(args.j_f_vf_m_ckpt)
else:
checkpoint = torch.load(args.j_f_vf_ckpt)
if len(modalities) == 2 :
if "vf" in modalities and "j" in modalities:
checkpoint = torch.load(args.v_j_ckpt)
if "vf" in modalities and "f" in modalities:
checkpoint = torch.load(args.v_f_ckpt)
print(1111)
if "f" in modalities and "j" in modalities:
checkpoint = torch.load(args.j_f_ckpt)
if len(modalities) == 1 :
if "vf" in modalities :
checkpoint = torch.load(args.v_ckpt)
if "f" in modalities :
checkpoint = torch.load(args.f_ckpt)
if "j" in modalities:
checkpoint = torch.load(args.j_ckpt)
state_dict = checkpoint['state_dict']
new_state_dict = {key.replace("actor.", "", 1): value for key, value in state_dict.items()}
loading_status = actor.load_state_dict(new_state_dict)
actor.eval()
plt.ion() # 启用Matplotlib的交互模式
try:
for _ in range(1):
ts = e.reset()
obs = ts.observation
# print("obs:",obs)
step = 0
reward_sum = 0
# if "vf" in modalities:
fig, ax1 = plt.subplots()
fig2, ax2 = plt.subplots()
image_brg = ts.observation['image']
stacked_images_rgb = []
for img_brg in image_brg:
img_rgb = cv2.cvtColor(img_brg, cv2.COLOR_BGR2RGB)
img_rgb = rearrange(img_rgb, 'h w c -> c h w')
resized_image = cv2.resize(img_rgb.transpose(1, 2, 0), (args.resized_height_v, args.resized_height_t)).transpose(2, 0, 1)
stacked_images_rgb.append(resized_image)
stacked_images_rgb = np.array(stacked_images_rgb)
while step < 500:
step += 1
print('step:',step)
# if "j" in modalities:
joint_numpy = np.array(ts.observation['joint'] / args.inspire)
joint_tensor = torch.from_numpy(joint_numpy).float().cuda()
joint_tensor = joint_tensor.unsqueeze(0)
force_numpy = np.array((ts.observation['force'] +1000)/2000 )
force_tensor = torch.from_numpy(force_numpy).float().cuda()
force_tensor = force_tensor.unsqueeze(0)
print('force_tensor',force_tensor)
images_tensor = torch.from_numpy(stacked_images_rgb / 255.0).float().cuda()
images_tensor = torch.unsqueeze(images_tensor, 0)
raw_action = actor(joint_tensor,force_tensor,images_tensor)
scores = raw_action[1]
print('scores_tensor',scores)
raw_action_tensor = raw_action[2]
force_tensor = raw_action[3]
force = force_tensor.cpu().detach().numpy()
force = force.squeeze()
raw_action_cpu = raw_action_tensor.cpu()
raw_action_numpy = raw_action_cpu.detach().numpy()
# ts = e.step(args,scores,raw_action_numpy,weight,force)
ts = e.step(args,scores,raw_action_numpy,force)
# ts = e.step(raw_action_numpy,weight)
obs = ts.observation
if "vf" in modalities:
image_data = obs['image'] # 获取更新后的图像数据
# image_data = cv2.cvtColor(image_data, cv2.COLOR_BGR2RGB)
# mask = np.zeros(image_data.shape[:2], dtype=np.uint8)
# r1 = (resized_height_t,resized_height_v, resized_width_t,resized_width_v)
# mask[r1[0]:r1[1], r1[2]:r1[3]] = 255
# ROI_image = cv2.bitwise_and(image_data, image_data, mask=mask)
# plt_img.set_data(ROI_image)
# plt.title("更新后的图像数据")
plt.pause(0.02)
# plt.draw() # 强制绘制
print(f'Total reward: {reward_sum}')
except KeyboardInterrupt:
print("强制退出循环")
finally:
# 清理部分
if 'actor' in locals():
del actor
torch.cuda.empty_cache()
plt.ioff()
plt.close('all')
e.reset()
if __name__ == '__main__':
import configargparse
p = configargparse.ArgParser()
p.add("-c", "--config", is_config_file=True, default="conf/imi/SoftGrasp_imi_visualize.yaml")
p.add("--batch_size", default=16)
p.add("--lr", default=1e-4, type=float)
p.add("--gamma", default=0.9, type=float)
p.add("--period", default=3)
p.add("--epochs", default=65, type=int)
p.add("--resume", default=None)
p.add("--num_workers", default=8, type=int)
p.add("--conv_bottleneck", required=True, type=int)
# p.add("--exp_name", required=True, type=str)
p.add("--encoder_dim", required=True, type=int)
p.add("--observation_dim", required=True, type=int)
p.add("--torque_dim", required=True, type=int)
p.add("--angle_dim", default=3, type=int)
p.add("--action_dim", default=3, type=int)
p.add("--share_dim", default=3, type=int)
p.add("--one_hot_torque_dim", default=3, type=int)
p.add("--one_hot_angle_dim", default=3, type=int)
p.add("--picture_dim", default=3, type=int)
p.add("--num_stack", required=True, type=int)
p.add("--frameskip", required=True, type=int)
p.add("--use_mha", default=True, action="store_true")
p.add("--use_amha", default=True, action="store_true")
p.add("--use_pos", required=True, type=int)
p.add("--use_one_hot", required=True, type=int)
p.add("--use_way", required=True)
p.add("--train_csv", default="data/train.csv")
p.add("--val_csv", default="data/val.csv")
p.add("--data_folder", default="data/test_recordings")
p.add("--env_name", default="hand-v1")
p.add("--resized_height_v", required=True, type=int)
p.add("--resized_width_v", required=True, type=int)
p.add("--resized_height_t", required=True, type=int)
p.add("--resized_width_t", required=True, type=int)
p.add("--num_episode", default=None, type=int)
p.add("--crop_percent", required=True, type=float)
p.add("--ablation", required=True)
p.add("--num_heads", required=True, type=int)
p.add("--encoder", required=True)
p.add("--use_flow", default=False, action="store_true")
p.add("--use_holebase", default=False, action="store_true")
p.add("--task", type=str)
p.add("--norm_audio", default=False, action="store_true")
p.add("--aux_multiplier", type=float)
p.add("--nocrop", default=False, type=int)
p.add("--inspire", default=1000, type=int)
## visualize
p.add("--weights_visualize_interval", required=True, type=int)
p.add("--weights_visualize", required=True, type=int)
p.add("--scores_visualize", required=True, type=int)
p.add("--save_scores", required=True, type=int)
p.add("--save_weights", required=True, type=int)
p.add("--camera_names", default="top")
p.add("--real_robot", default="top")
p.add("--arm", default="top")
p.add("--ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--f_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--v_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--j_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--v_j_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--v_f_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--j_f_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--j_f_vf_m_pos_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--j_f_vf_m_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
p.add("--j_f_vf_ckpt", default="04-29-11:12:31-jobid=0-epoch=0-step=325.ckpt")
# transformer
p.add("--hidden_dim", required=True, type=int)
p.add("--dropout", required=True, type=float)
p.add("--nheads", required=True, type=int)
p.add("--dim_feedforward", required=True, type=int)
p.add("--enc_layers", required=True, type=int)
p.add("--dec_layers", required=True, type=int)
p.add("--pre_norm", default=True, action="store_true")
args = p.parse_args()
args.batch_size *= torch.cuda.device_count()
main(args)