help me, I get an error when I run run-cluster.sh
i run :
./bin/run-cluster.sh 1:2:1 --train_files /home/dev/ds041/DeepSpeech/Data/train/train2.csv --dev_files /home/dev/ds041/DeepSpeech/Data/dev/dev2.csv --test_files /home/dev/ds041/DeepSpeech/Data/test/test2.csv --train_batch_size 1 --dev_batch_size 1 --test_batch_size 1 --n_hidden 494 --epoch 5 --checkpoint_dir “$checkpoint_dir” --alphabet_config_path /home/dev/ds041/Indonesia-models/alphabet.txt --lm_binary_path /home/dev/ds041/Indonesia-models/lm.binary --lm_trie_path /home/dev/ds041/Indonesia-models/trie
and i got error :
InvalidArgumentError (see above for traceback): Restoring from checkpoint failed. This is most likely due to a mismatch between the current graph and the graph from the checkpoint. Please ensure that you have not altered the graph expected based on the checkpoint. Original error:
[ps 0]
[ps 0] Cannot assign a device for operation fifo_queue_enqueue: Could not satisfy explicit device specification ‘’ because the node node fifo_queue_enqueue (defined at /home/dev/ds041/DeepSpeech/util/config.py:127) having device No device assignments were active during op ‘fifo_queue_enqueue’ creation. was colocated with a group of nodes that required incompatible device ‘/job:ps/task:0’
[ps 0] Colocation Debug Info:
[ps 0] Colocation group had the following types and devices:
[ps 0] QueueDequeueV2: CPU XLA_CPU XLA_GPU
[ps 0] QueueEnqueueV2: CPU XLA_CPU XLA_GPU
[ps 0] FIFOQueueV2: CPU XLA_CPU XLA_GPU
[ps 0]
[ps 0] Colocation members and user-requested devices:
[ps 0] fifo_queue (FIFOQueueV2) /job:ps/task:0
[ps 0] fifo_queue_enqueue (QueueEnqueueV2)
[ps 0] fifo_queue_Dequeue (QueueDequeueV2)
[ps 0]
[ps 0] [[node fifo_queue_enqueue (defined at /home/dev/ds041/DeepSpeech/util/config.py:127) = QueueEnqueueV2[Tcomponents=[DT_INT32], timeout_ms=-1](fifo_queue, Placeholder)]]
[ps 0]
[ps 0] No node-device colocations were active during op ‘fifo_queue_enqueue’ creation.
[ps 0] No device assignments were active during op ‘fifo_queue_enqueue’ creation.