jhu-lcsr/good_robot

"Good Robot! Now Watch This!": Repurposing Reinforcement Learning for Task-to-Task Transfer; and “Good Robot!”: Efficient Reinforcement Learning for Multi-Step Visual Tasks with Sim to Real Transfer

good_robot's Issues

Why the generated real robot‘s color heightmap is not good？？？

Hello, first of all thanks for sharing your amazing job. I want to know Why the generated real robot‘s color heightmap is not good？My color heightmaps are often deformed and have a lot of black Points. Is this caused by inaccurate camera calibration? I have try with VPG‘s calibration method.(0.448x0.448 resolution=0.02 and 0.672x0.672 resolution=0.03)
and ROS easy_handeye method. the second method cant get the same R.t matrix with the first. so i can‘t use it.

Thanks in advance ;)

Detect Changes

Hi Dear;
I was looking at your code, I found this part Detect Changes
I need help to understand what does it mean 0.3 , 0.01, et?. what is the standard of 300 and how it is obtained to be as change_threshold?.
May anyone help me in clarification this part, please?.
I'd appreciate your cooperation

`if 'prev_color_img' in locals():

        # Detect changes
        depth_diff = abs(depth_heightmap - prev_depth_heightmap)
        depth_diff[np.isnan(depth_diff)] = 0
        depth_diff[depth_diff > 0.3] = 0
        depth_diff[depth_diff < 0.01] = 0
        depth_diff[depth_diff > 0] = 1
        change_threshold = 300
        change_value = np.sum(depth_diff)
        change_detected = change_value > change_threshold or prev_grasp_success  
        print('Change detected: %r (value: %d)' % (change_detected, change_value))`

Sincerely

issues when loading pretrained models for testing

Hi Andrew,

When I was using the pretrained models for testing, the robot didn't behave as expected in stacking and row. I think I try to follow the guides to reproduce the results. I record the performance in the video, can you please take a look and let me know where I might be wrong? I am using the master branch.

https://drive.google.com/open?id=17hiBGqGT4eqgtK9kSCwHBNZxwGgixtLg

Thank you in advance!

Best,
Yang

RuntimeError: CUDA out of memory. Tried to allocate 20.00 MiB (GPU 0; 7.79 GiB total capacity; 6.03 GiB already allocated; 10.94 MiB free; 6.17 GiB reserved in total by PyTorch)

I tried to follow this repository. However, when I entered

export CUDA_VISIBLE_DEVICES="0" && python3 main.py --is_sim --obj_mesh_dir objects/blocks --num_obj 4 --push_rewards --experience_replay --explore_rate_decay --check_row --tcp_port 19997 --place --future_reward_discount 0.65 --max_train_actions 20000 --random_actions --common_sense --trial_reward

I got following error

Traceback (most recent call last):
  File "main.py", line 1755, in <module>
    one_train_test_run(args)
  File "main.py", line 1563, in one_train_test_run
    training_base_directory, best_dict = main(args)
  File "main.py", line 1141, in main
    trainer.backprop(prev_color_heightmap, prev_valid_depth_heightmap, prev_primitive_action, prev_best_pix_ind, label_value, goal_condition=prev_goal_condition)
  File "/home/khan/cop_ws/src/good_robot/trainer.py", line 717, in backprop
    push_predictions, grasp_predictions, place_predictions, state_feat, output_prob = self.forward(color_heightmap, depth_heightmap, is_volatile=False, specific_rotation=best_pix_ind[0], goal_condition=goal_condition)
  File "/home/khan/cop_ws/src/good_robot/trainer.py", line 445, in forward
    output_prob, state_feat = self.model.forward(input_color_data, input_depth_data, is_volatile, specific_rotation, goal_condition=goal_condition)
  File "/home/khan/cop_ws/src/good_robot/models.py", line 246, in forward
    interm_push_feat, interm_grasp_feat, interm_place_feat, tiled_goal_condition = self.layers_forward(rotate_theta, input_color_data, input_depth_data, goal_condition, tiled_goal_condition)
  File "/home/khan/cop_ws/src/good_robot/models.py", line 301, in layers_forward
    interm_place_depth_feat = self.place_depth_trunk.features(rotate_depth)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/container.py", line 119, in forward
    input = module(input)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/container.py", line 119, in forward
    input = module(input)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/khan/anaconda3/envs/raisim_env/lib/python3.8/site-packages/torchvision/models/densenet.py", line 33, in forward
    new_features = super(_DenseLayer, self).forward(x)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/container.py", line 119, in forward
    input = module(input)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/modules/batchnorm.py", line 135, in forward
    return F.batch_norm(
  File "/home/khan/.local/lib/python3.8/site-packages/torch/nn/functional.py", line 2149, in batch_norm
    return torch.batch_norm(
RuntimeError: CUDA out of memory. Tried to allocate 2.00 MiB (GPU 0; 7.79 GiB total capacity; 6.09 GiB already allocated; 28.69 MiB free; 6.26 GiB reserved in total by PyTorch)

After reading some blogs, I found some discussion here which pointed out that it is due to batch size and solution is possible if batch size is reduced. To do this, I tried trainer.py file and I found following section of the code

# Construct minibatch of size 1 (b,c,h,w)
        input_color_image.shape = (input_color_image.shape[0], input_color_image.shape[1], input_color_image.shape[2], 1)
        input_depth_image.shape = (input_depth_image.shape[0], input_depth_image.shape[1], input_depth_image.shape[2], 1)
        input_color_data = torch.from_numpy(input_color_image.astype(np.float32)).permute(3,2,0,1)
        input_depth_data = torch.from_numpy(input_depth_image.astype(np.float32)).permute(3,2,0,1)
        if self.flops:
            # sorry for the super random code here, but this is where we will check the
            # floating point operations (flops) counts and parameters counts for now...
            print('input_color_data trainer: ' + str(input_color_data.size()))
            class Wrapper(object):
                custom_params = {'input_color_data': input_color_data, 'input_depth_data': input_depth_data, 'goal_condition': goal_condition}
            def input_constructor(shape):
                return Wrapper.custom_params
            flops, params = get_model_complexity_info(self.model, color_heightmap.shape, as_strings=True, print_per_layer_stat=True, input_constructor=input_constructor)
            print('flops: ' + flops + ' params: ' + params)
            exit(0)
        # Pass input data through model
        output_prob, state_feat = self.model.forward(input_color_data, input_depth_data, is_volatile, specific_rotation, goal_condition=goal_condition)

I think I need to divide something to reduce batch size. Can anyone help please?

--batch-size

How to modify --batch-size？
Thank you！