1 | git log |
1 | commit 0637340f4d8d288a8d24dd5193855c9e5def776d (HEAD -> main, origin/main, origin/HEAD) |
1 | git reset --hard [version hash] |
1 | git pull |
or locally
1 | git checkout master |
1 | git clone ... |
1 | git log |
A Git submodule allows a user to include a Git repository as a subdirectory of another Git repository. This can be useful when a project needs to include and use another project. For example, it may be a third-party library or a library developed independently for use in multiple parent projects. With submodules, these libraries can be managed as independent projects while being used in the user’s project. This allows for better organization and management of the code.
To add an existing Git repository as a submodule to a project, the git submodule add command can be used. The command format is git submodule add <url> <path>, where <url> is the URL of the submodule repository and <path> is the storage path of the submodule in the project. For example, if a user wants to add the remote repository https://github.com/username/repo.git as a submodule to their project and store it in the my-submodule directory, they can use the following command:
1 | git submodule add https://github.com/username/repo.git my-submodule |
This will successfully add a submodule named my-submodule to the user’s project.
raw data : for each subjects(S1,S2 …) , each action(walking, waiting, smoking …), each sub sequence(1/2):
$(n) \times 99$ (np.ndarray, float32)
data_utils.load_data() used by translate.read_all_data()train data: the composed dictionary ((suject_id, action, subaction_id, ‘even’) as key) of raw data (just even rows), with one hot encoding columns for action type, if action is specified (normal case), just append an all 1 column to rawdata. Size of each dictionary value:
$(n/2) \times (99 + actions;count)$
complete data: all data joint together, from different subjects, actions, sub sequences:
$(n) \times 99$
translate.read_all_data() used by translate.train()train set : normalized train data, throw out data with $std < 1e-4$ (accroding to complete data). Size of each dictionary value:
$(n/2) \times ((99-used;dimension;count) + actions;count)$
After the analyzztion of the complete data, human dimension has been fixed to $54$.
Seq2SeqModel.get_batch() used by translate.train()total_seq: $60$ ($[0,59]$)
source_seq_len: $50$
target_seq_len: $10$
batch_size: $16$
encoder_inputs: $16\times 49\times (54+actions;count)$
Interpretation: [batch,frame,dimension]
frame range: $[0,48]$
decoder_inputs: $16\times 10\times (54+actions;count)$
frame range: $[49,58]$
decoder_outputs: $16\times 10\times (54+actions;count)$
frame range: $[50,59]$
encoder_inputs: Tensor form of encoder_inputs from Seq2SeqModel.get_batch()
1 | torch.from_numpy(encoder_inputs).float() |
decoder_inputs: Tensor form of decoder_inputs from Seq2SeqModel.get_batch()
For detailed usage, please see [Adopted] human-motion-prediction-pytorch\src\predict.ipynb
The kinect camera’s output is not guaranteed to be consistent with the input of this model (some features are cut off), so further research is needed.
Run pyKinectAzure\examples\exampleBodyTrackingTransformationComparison to get the camera output record in pyKinectAzure\saved_data, saved as .npy
link
ALWAYS
DONE!
WORKING~
Deep-Reinforcement-Learning-With-Python
In supervised learning, the machine learns from training data. The training data consists of a labeled pair of inputs and outputs. So, we train the model (agent) using the training data in such a way that the model can generalize its learning to new unseen data. It is called supervised learning because the training data acts as a supervisor, since it has a labeled pair of inputs and outputs, and it guides the model in learning the given task.
Quantitative response
predict a quantitative variable from a set of features
Categorical response
predict a categorical variable
Similar to supervised learning, in unsupervised learning, we train the model (agent) based on the training data. But in the case of unsupervised learning, the training data does not contain any labels; that is, it consists of only inputs and not outputs. The goal of unsupervised learning is to determine hidden patterns in the input. There is a common misconception that RL is a kind of unsupervised learning, but it is not. In unsupervised learning, the model learns the hidden structure, whereas, in RL, the model learns by maximizing the reward.
The set of all possible actions in the environment is called the action space. Thus, for this grid world environment, the action space will be [up, down, left, right]. We can categorize action spaces into two types:
A policy defines the agent’s behavior in an environment. The policy tells the agent what action to perform in each state.
Over a series of iterations, the agent will learn a good policy that gives a positive reward.
The optimal policy tells the agent to perform the correct action in each state so that the agent can receive a good reward.
Deterministic Policy
deterministic policy tells the agent to perform a one particular action in a state. Thus, the deterministic policy maps the state to one particular action
Stochastic Policy
maps the state to a probability distribution over an action space.
The agent interacts with the environment by performing some action starting from the initial state and reach the final state. This agent-environment interaction starting from the initial state until the final state is called an episode. For instance, in the car racing video game, the agent plays the game by starting from the initial state (starting point of the race) and reach the final state (endpoint of the race). This is considered an episode. An episode is also often called trajectory (path taken by the agent)
Horizon is the time step until which the agent interacts with the environment. We can classify the horizon into two:
Return is the sum of rewards received by the agent in an episode.
Value function or the value of the state is the expected return that the agent would get starting from the state $s$ following the policy $\pi$
implies the expected return agent would obtain starting from the state $s$ and an action $a$ following the policy $\pi$.
Prime theory: An integrated view of motivation and emotion
what is my motivation
Intrinsic Motivation and norms about payment
what is my emotion
In old days people have high spirits
the obs dictionary’s builder
For me, intrinsic and extrinsic motivation differs a lot. Since my early age, I distinguish clearly between schoolwork and extracurricular activities. During week days, I devote 100% of my energy into schoolwork and during weekends, I participate in robot and model aircraft clubs.
Before I entered college, I think that completing coursework is a student’s duty and it is also very rewarding. By behaving well in school, I can receive praise from my parents, respectation from peers, promising jobs and For this reason, completing schoolwork is mainly motivated by extrinsic rewards.
My dad used to have an argue with my mum about whether they should reward me by giving me some money to buy what I want if I perform well in exams. He thought exam scores should not be related with money rewards. While my mum thought it is reasonable since rewards can motivate me a lot. After I entered college, I thought my dad was right, although money rewards did motivate me to get higher scores, it is extrinsic motivation. According to the textbook, extrinsic rewards can reduce intrinsic motivation. When I look back, I cannot clearly figure out what is my intrinsic motivation of studying hard at school. Most of the time, I am motivated to study a subject just for the reason that I am good at it. If I am not good at a subject, such as Chinese, I just keep persuding me that if my Chinese is not learnt well, I will not be able to enter a good school. Usually, during this process, I suffer from heavy depression and anxiety. Speaking of how much passion I have in all of these subjects, I have to say, very little. My behaviour in exam-oriented education is like a machine. I get rewards, commitments as inputs and output good scores.
Although I behaves like the walking dead in schoolwork, I have intrinsic motivations in other fields, such as building robots and flying model aircraft. These activities fulfill no obvious purpose other than enjoyment.
What affects your health?
Biological factors refer to innate or naturally occurring factors such as genetics, age, and gender that can affect your physical and mental health.
Psychological factors refer to factors related to your thoughts and emotions such as emotional state, cognitive processes, and behavioral patterns that can affect your physical and mental health.
Social factors refer to environmental factors such as culture, family, social networks that can affect your physical and mental health.
Stress refers to tension, anxiety or challenge from internal or external environment which may have negative effects on both physical and mental health.
Mediating factors are factors that can help alleviate the negative effects of stress. These factors include positive coping strategies, seeking social support, and changing cognitive processes.
A positive attitude has positive effects on both your physical and mental health. It can help you cope with stress, improve your immune system, and increase your overall well-being.
This section provides tips and suggestions for maintaining physical and mental health. It includes advice on diet, exercise, sleep, relaxation techniques, social networks, and seeking support.
Take the movie “Les Choristes” as an example
Principals stand for Behaviorism:
In the face of confused or lost teenagers, the principal’s education method is very representative of the indifferent and rigid “behaviorist”. He believes that violence and punishment can make children keep their duties. The endless and varied punishment methods are his masterpieces. In the face of the principal’s violence, the children really behaved in silence and followed the rules, but they did not believe the principal from the bottom of their hearts. Once the principal leaves and doesn’t pay attention, the children will play tricks behind their backs and vent their emotions.
One of the core viewpoints of behaviorism is that environment is the only condition for the learning, and learning is completely determined by the external environment. Therefore, the principal puts the focus of education on corporal punishment, punishment, reprimand, and criticism. Even if children make small mistakes, they will always be scolded or beaten.
It is true that punishment as an external stimulus can promote children to develop in a good way to a certain extent, but the fear generated by punishment is often greater than the education gained, so children will become more and more disobedient, and they will not learn what they should know at all.
When we learn that a stimulus predicts another stimulus.
When we learn that a behavior leads to a certain outcome.
When we learn or change a behavior after watching a person engage in that behavior
Displaying a behavior that imitates a previously observed behavior
learning about an
action’s consequences by observing others being rewarded or
punished for their behavior
Mathews stands for Humanism:
Compared with the principal’s style, Teacher Matthew regards the children as himself, understands their distress, and walks into their hearts. Teacher Matthew is the first to really treat “problem teenagers” as dignified and thoughtful individuals. Teacher Matthew’s education and enlightenment are the classic characteristics of humanists.
Humanism holds that different people will have different beliefs about the same fact, which means different things to different people. Therefore, humanists can always start from the individual’s inner feelings and beliefs, and touch others with their feelings and ideas.
Teacher Matthew pays attention to discovering children’s potential in learning, and constantly encourages children to improve and cheer. Combined with rationality, it only helps children to affirm themselves and realize themselves.
