Fix error in custom Gym environment för reinforced learning with Python
$10-30 USD
Completed
Posted over 1 year ago
$10-30 USD
Paid on delivery
I have a Python program with a custom Open AI Gym environment.
After maybe 50 000 – 200 000 time steps I get an error.
I have tried on different computers and I tried to reinstall Anaconda environment but the problem is still there. (So I can make the error happen on multiple computers)
I want help finding and fixing that problem so the program can run for millions of time steps.
Error:
------------------------------------
| time/ | |
| fps | 258 |
| iterations | 25300 |
| time_elapsed | 489 |
| total_timesteps | 126500 |
| train/ | |
| entropy_loss | -0.054 |
| explained_variance | 0.0318 |
| learning_rate | 0.0007 |
| n_updates | 25299 |
| policy_loss | 29.4 |
| value_loss | 2.14e+07 |
------------------------------------
Traceback (most recent call last):
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\spyder_kernels\[login to view URL]", line 356, in compat_exec
exec(code, globals, locals)
File "c:\_sb3withoanda\[login to view URL]", line 717, in <module>
[login to view URL](total_timesteps=nSteps)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\stable_baselines3\a2c\[login to view URL]", line 191, in learn
return super(A2C, self).learn(
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\stable_baselines3\common\[login to view URL]", line 250, in learn
continue_training = self.collect_rollouts([login to view URL], callback, self.rollout_buffer, n_rollout_steps=self.n_steps)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\stable_baselines3\common\[login to view URL]", line 169, in collect_rollouts
actions, values, log_probs = [login to view URL](obs_tensor)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\stable_baselines3\common\[login to view URL]", line 592, in forward
distribution = self._get_action_dist_from_latent(latent_pi)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\stable_baselines3\common\[login to view URL]", line 610, in _get_action_dist_from_latent
return self.action_dist.proba_distribution(action_logits=mean_actions)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\stable_baselines3\common\[login to view URL]", line 274, in proba_distribution
[login to view URL] = Categorical(logits=action_logits)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\torch\distributions\[login to view URL]", line 66, in __init__
super(Categorical, self).__init__(batch_shape, validate_args=validate_args)
File "C:\anaconda3\envs\GymSB3Oanda\lib\site-packages\torch\distributions\[login to view URL]", line 56, in __init__
raise ValueError(
ValueError: Expected parameter logits (Tensor of shape (1, 3)) of distribution Categorical(logits: [login to view URL]([1, 3])) to satisfy the constraint IndependentConstraint(Real(), 1), but found invalid values:
tensor([[nan, nan, nan]], device='cuda:0')
Dear client,
I am a senior software engineer who has vast experience with tech stacks below for 12 years.
- Tradition Machine Learning Algorithm with Scikit-Learn
- Tiktok, YOLO v7 for Image & Video Detection
-Deep Neural Networks Analtsis
- Data Structures - manage the database using mongo or MySql
- NLP with GPT-3(Text-to-Speech, Vice Versa)
- Computer Vision(OpenCV)
- PyTorch
- Tensorflow and OpenAI
- OCR
In here, my major skill is NLP, have full experience in signal processing
I'm familiar with agile project management tools including Slack, JIRA, Trello, Bitbucket, Github, etc.
I ensure the highest quality of product and 100% satisfaction through my work.
I can work on 7/24 for your project.
I am innovative and strategic thinking professional with a proven track record of consistently going above and beyond in meeting customer needs and providing more value to the product than what the customer is paying for.
For this very reason, they always get back to us again and again with promising ideas and projects.
I wish we can discuss more details in chat.
I'll look forward to hearing from you soon.
Thanks so much.
Kind Regards.
$20 USD in 7 days
3.3 (2 reviews)
1.7
1.7
4 freelancers are bidding on average $20 USD for this job
Hello sir how are you doing? I have read the project detail and really interested in your project, I am full stack developer with multiple frameworks, I have great experience doing similar jobs regarding to these skills Open AI, Python and Troubleshooting.
Please start the chat, also I have some questions so we can have detailed discussion about project and finalize the timeline. Thanks
Regards
Umair
I have an experienced python team, I trust that my team can deliver all of your requirements on time with the desired quality. Hire us and you will not regret it.
Hi!
I reade your requirements you need a good designer who will create aunique style,pattern and good information.
your reading right proposal i wil help you in short interval of time .
i will use in your logo these 5 principle that will make your logo most effective and catch eye.
Simple. Your logo needs to be easily identifiable at a glance. ...
Memorable. An effective logo should be memorable. ...
Timeless. An effective logo should be timeless and should avoid trends. ...
Versatile. A good logo can be used in a variety of sizes and colours. ...
Appropriate.
here is some special service which i will provide you.
1:24/5 Avalibility
2: unlimited reavision
3: Design in just 2 hours
4:Adobe Illustrator (AI)
5:Editable PDF.
6:EPS.
7:SVG.
8:JPEG.
9:PNG.
i hope you like my proposal i will wait for your positive response
is there any Query ask me freely Thanks.