"Couldn't match files for checkpoint" when trying to visualize embedding

Hi,

I am trying to visualize an embedding with the tensorboard projector. I save the checkpoint files like instructed in the tensorflow tutorial and they are located in the projectfolder/logs/checkpointFFNtest.ckpt
When running Tensorboard it gives me this error: Couldn't match files for checkpoint .//dropOutFFNTest.ckpt
The checkpoint file is definitely present and can also be read into tensorflow with
saver.restore(sess, "C:/MyFiles/Uni/Tensorflow/project1/log/dropOutFFN.ckpt")
Tensorboard also finds the scalars which are saved in C:/MyFiles/Uni/Tensorflow/project1/summaries/train(or /val) and visualizes them correctly.
I've tried to show embeddings with the projector by starting tensorboard in the overall project directory as well as directly in the "logs" directory. Do you know what the problem might be?

Here a condensed code snipped:

from tensorflow.contrib.tensorboard.plugins import projector
tf.reset_default_graph()
%matplotlib notebook

X = tf.placeholder("float", [None,n_input])
Y = tf.placeholder("float",[None,n_classes])
keep_probability = tf.placeholder("float")

weights = {
'h1': tf.Variable(tf.random_normal([n_input, n_hidden_1])),
'h2': tf.Variable(tf.random_normal([n_hidden_1, n_hidden_2])),
'out': tf.Variable(tf.random_normal([n_hidden_2, n_classes]))
}

biases = {
'b1': tf.Variable(tf.random_normal([n_hidden_1])),
'b2': tf.Variable(tf.random_normal([n_hidden_2])),
'out': tf.Variable(tf.random_normal([n_classes]))
}
def multilayer_perceptron(x):
# Hidden fully connected layer with 256 neurons
layer_1 = tf.nn.dropout(tf.nn.tanh(tf.add(tf.matmul(x, weights['h1']), biases['b1'])),keep_probability)
# Hidden fully connected layer with 256 neurons

layer_2 = tf.nn.dropout(tf.nn.tanh(tf.add(tf.matmul(layer_1, weights['h2']), biases['b2'])),keep_probability)
# Output fully connected layer with a neuron for each class
out_layer = tf.matmul(layer_2, weights['out']) + biases['out']
return out_layer

out = multilayer_perceptron(X)

predicted=tf.sigmoid(out)
[...]
loss_op = tf.reduce_mean(tf.nn.sigmoid_cross_entropy_with_logits(
logits=out, labels=Y))#sigmoid because classes are not mutially exclusive
optimizer = tf.train.AdamOptimizer(learning_rate=learning_rate)
train_op = optimizer.minimize(loss_op)

tf.summary.scalar("cross_entropy", loss_op)
tf.summary.scalar("accuracy",accuracy)
tf.summary.scalar("NumRight",tf.reduce_mean(NumRightPred))
tf.summary.scalar("NumWrong",tf.reduce_mean(NumFalsePred))
tf.summary.scalar("absDiff",tf.reduce_mean(diff))

merged_summaries = tf.summary.merge_all()
init = tf.global_variables_initializer()

"""summary_writer = tf.summary.FileWriter("./summaries/new")
config = projector.ProjectorConfig()
embedding = config.embeddings.add()
embedding.tensor_name = weights['h1'].name
projector.visualize_embeddings(summary_writer, config)"""#tried this but it didn't work

train_writer = tf.summary.FileWriter("./summaries/train", tf.get_default_graph())
validation_writer = tf.summary.FileWriter("./summaries/validation")
with tf.Session() as sess:
#sess.run(tf.global_variables_initializer())
sess.run(init)
saver = tf.train.Saver()

saver.restore(sess, "C:/MyFiles/Uni/Tensorflow/project1/log/dropOutFFN.ckpt")
print("model loaded")
for n in range(training_epochs):   
    _,loss,predTreat,_summaries,acc,absdiff = sess.run([...], feed_dict={...})             

    train_writer.add_summary(_summaries, step)
    if (step % 50 == 0): 
        kp=1.0
        icd, treatment = getBatch(batch_size,"val")
        _,loss_val,predTreat,_summaries,acc,absdiff = sess.run([...], feed_dict={...})
        print("Accuracy: ",acc)
        print("Cross Entropy: ",loss_val)
        print("Diff: ",absdiff)
        validation_writer.add_summary(_summaries, step)
    if (step%1000==0):
        saver.save(sess,"./log/dropOutFFNTest.ckpt")   
    step = step + 1