Cannon run example code: Traceback (most recent call last): · Issue #498

link管理

链接快照平台

输入网页链接，自动生成快照
标签化管理网页链接

相关文章推荐

微醺的红茶 · 商演拒绝假唱被称赞！魏晨晒图谦虚回应：这是我 ...· 1 年前 ·

成熟的小狗 · Stormy Daniels Shares ...· 1 年前 ·

深情的乌冬面 · 2023年2月烤箱推荐家用|烤箱可以做什么美 ...· 1 年前 ·

有腹肌的熊猫 · 两人跳棋开局6步必胜_头条· 1 年前 ·

跑龙套的青蛙 · 泛突厥主义- 知乎· 1 年前 ·

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement . We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

I'm using Mac and I'm trying out gym from openai, then I downloaded a project for which I needed stable-baselines so I followed the documentation and it didn't work so I thought maybe it's gym but gym works fine. I made a new project copy and pasted the example code on stable-baselines documentation, but it just keeps giving me the same error. I tried every way of downloading stable-baseline: pip, pip3, with and without mpi, I even tried downloading it via github and the bleeding-edge version. I already have CMake and openmpi

Code:

import gym
from stable_baselines.common.policies import MlpPolicy
from stable_baselines.common.vec_env import DummyVecEnv
from stable_baselines import PPO2
env = gym.make('CartPole-v1')
env = DummyVecEnv([lambda: env])  # The algorithms require a vectorized environment to run
model = PPO2(MlpPolicy, env, verbose=1)
model.learn(total_timesteps=10000)
obs = env.reset()
for i in range(1000):
    action, _states = model.predict(obs)
    obs, rewards, dones, info = env.step(action)
    env.render()

Error:

Traceback (most recent call last):
  File "/Users/(my name)/Documents/testing gym.py", line 3, in <module>
    from stable_baselines.common.policies import MlpPolicy
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines/__init__.py", line 4, in <module>
    from stable_baselines.deepq import DQN
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines/deepq/__init__.py", line 1, in <module>
    from stable_baselines.deepq.policies import MlpPolicy, CnnPolicy, LnMlpPolicy, LnCnnPolicy
  File "/Library/Frameworks/Python.framework/Versions/3.7/lib/python3.7/site-packages/stable_baselines/deepq/policies.py", line 2, in <module>
    import tensorflow.contrib.layers as tf_layers
ModuleNotFoundError: No module named 'tensorflow.contrib'
System Info

Describe the characteristic of your environment:
I installed via pip(3), GitHub clone
I'm using a Mac with intel core i7 quad-core 2.6GHz and Radeo Pro 560X 4gb
My python version is python 2.7.11, python3 3.7.4
Tensorflow version: 2.0.0
gym version: 0.14.0
Please help
Tensorflow version: 2.0.0
That there is the problem: Stable-baselines is built for Tensorflow major version 1 (up till 1.14.0), and does not work for Tensorflow 2. You can change to working tensorflow version with pip3 install --upgrade tensorflow==1.14.0.
I think this is a problem in stable-baselines as setup.py attempts to install any tensorflow version up from 1.8.0, including 2.0.0.
      When I ran the "Getting started" example I have the following error: "ModuleNotFoundError: No module named 'tensorflow.contrib'."
it seems there is tensorflow older than 2.1 is not available any more (at least on python 3.9) what should I do?

tanks