search

PathologyDataScience / HistomicsML2

A tool for training machine-learning models with whole-slide imaging datasets
22 stars 5 forks source link

Could not import MapnikTileSource #47

Closed Acmenwangtuo closed 4 years ago

Acmenwangtuo commented 4 years ago

When I click review,i can't see the result Screenshot from 2020-02-04 14-41-34

Acmenwangtuo commented 4 years ago

Screenshot from 2020-02-04 14-49-36

slee172 commented 4 years ago

I think one of the reasons of the error is that your ip address of the server container is not properly set up. After starting the Apache server "service apache2 start", you need to set up the ip address here "/var/www/html/HistomicsML/php/hostspecs.php". Confirm it using "cat /var/www/html/HistomicsML/php/hostspecs.php", then you can see "$IIPServer = "http://localhost/fcgi-bin/iipsrv.fcgi?";". The IIPServer setting should be "localhost" like above if you don't use static IP address on your own machine (i.e. laptop). However, the IIPServer setting should be your IP address (i.e. xxx.xxx.xxx.xxx) if your machine has been assigned static IP address (This typically happens when you are using a server). Another reason I would guess is that your database container is not properly connected to the server container. Try to find the docker container for the database. See the subsection "Runing the HistomicsML database container" from https://histomicsml-taf.readthedocs.io/en/latest/example-data.html If you still see the error, showing your full log message would be helpful.

cooperlab commented 4 years ago

Fixed by pull request #48

Acmenwangtuo commented 4 years ago

It still don't work INFO:root:Created LRU Cache for 'tilesource' with 2686 maximum size INFO:root:Notice: Could not import MapnikTileSource Using TensorFlow backend. WARNING: Logging before flag parsing goes to stderr. W0303 07:05:39.299856 140228533643072 deprecation_wrapper.py:119] From /usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py:74: The name tf.get_default_graph is deprecated. Please use tf.compat.v1.get_default_graph instead.

W0303 07:05:39.314635 140228533643072 deprecation_wrapper.py:119] From /usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py:517: The name tf.placeholder is deprecated. Please use tf.compat.v1.placeholder instead.

W0303 07:05:39.316373 140228533643072 deprecation_wrapper.py:119] From /usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py:4138: The name tf.random_uniform is deprecated. Please use tf.random.uniform instead.

W0303 07:05:39.338979 140228533643072 deprecation_wrapper.py:119] From /usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py:3976: The name tf.nn.max_pool is deprecated. Please use tf.nn.max_pool2d instead.

W0303 07:05:40.866888 140228533643072 deprecation_wrapper.py:119] From /usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py:174: The name tf.get_default_session is deprecated. Please use tf.compat.v1.get_default_session instead.

W0303 07:05:40.867279 140228533643072 deprecation_wrapper.py:119] From /usr/local/lib/python2.7/dist-packages/keras/backend/tensorflow_backend.py:181: The name tf.ConfigProto is deprecated. Please use tf.compat.v1.ConfigProto instead.

Acmenwangtuo commented 4 years ago

image

slee172 commented 4 years ago

@Acmenwangtuo I think this issue should be closed. You are supposed to do just re-loaded one of pre-trained sample classifiers rather than creating use one. This means that the whole-slide information should not be displayed in the Review because you ran the example docker that doesn't have all whole-slides (I think). If you get a change to confirm that the Review tab works, you probably need to create a new classifier and then go and confirm the Review after choosing new samples.