'ResourceExhaustedError: Exception encountered when calling layer "batch_normalization_10" (type BatchNormalization)
When I run a transformer code for video prediction, the code itself should be correct but an error occured:
ResourceExhaustedError: Exception encountered when calling layer "batch_normalization_10" (type BatchNormalization).
OOM when allocating tensor with shape[8,4,1600,64] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc [Op:FusedBatchNormV3]
Call arguments received: • inputs=tf.Tensor(shape=(8, 4, 40, 40, 64), dtype=float32) • training=True
How can I tackle this problem? I think it is not a very big tensor...
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|