Open NamburiSrinath opened 5 years ago
Hi Srinath,
example_1 and example_2 are easy to understand, but will often produce really bad results. example_3 and example_4 are much better (even for single object recognition) since they are based on actual papers. For example, I used example_1 for OCR and also got bad results but with example_4 I obtained >85% IoU.
If you still want to use example_1, try setting trainable=True and/or adding some Conv2D layers. The screenshot you sent is actually showing that the network is improving. You can see that val_mse is initially 141 and then goes down to 120, but the improvements are not big enough.
If you configured TensorFlow correctly, the GPU will be used automatically. Check tf.test.is_gpu_available()
.
-- Lars
Hi Lars,
I have tried running the code on a dataset for 100 epochs. After around 50 epochs, the error reduced from 2000(initially) to 25 but it stopped suddenly. How many times I have tried, I end up with same result. I couldn't figure out why it is stopping in between.
Could you help me in resolving this issue?
The exact error is placed in a screenshot. It is saying the val_iou didn't improve from 0.
Also, how to use GPU to run the python script. Currently it has multi-threading options which use CPU but I have a GPU supported system. How to change the access?
Thanks in advance lars Srinath