search

naokishibuya / car-behavioral-cloning

Built and trained a convolutional network for end-to-end driving in a simulator using Tensorflow and Keras
MIT License
230 stars 136 forks source link

The model.h5 is not working as expected. #5

Closed chaitanyya closed 7 years ago

chaitanyya commented 7 years ago

We tried to run the model.h5 file and for some reason, it always crashes at the first turn in the jungle track. Is it the model you used in the video?

naokishibuya commented 7 years ago

Yes, the file was used for the video. One possible cause for the crash might be the library version differences. environment.yml or environment-gpu.yml has the conda environment details. Let me know if you still have the issue even though the environment is the same.

xeron56 commented 6 years ago

Same problem i faceed; i installed all package manually with pip and the model don't work

@chaitanyya you can try this

First you need to

  1. open anaconda promt as adminstrator
  2. write conda env list ferrthrth

see that if the previous env with name "car-behavioral-cloning" has existed there then delete it

with conda remove --name car-behavioral-cloning --all

  1. Now wef

write anaconda login for log into the anaconda server, give username and pass , or no package will be downloaded

  1. write cd /d H:\How_to_simulate_a_self_driving_car here place your won gitclone directory

  2. Now Write conda env create -f environment.yml

  3. activate the new env with activate car-behavioral-cloning

dwqwdq

  1. Now write python drive.py model.h5

thanks @naokishibuya code works superbly fine ..