sorry about the post i have figured it out, it was a typo issue that i missed. i created my own tf,record.
but i do get this error
InvalidArgumentError (see above for traceback): logits and labels must have the same first dimension, got logits shape [4,3] and labels shape [1]
[[node SparseSoftmaxCrossEntropyWithLogits/SparseSoftmaxCrossEntropyWithLogits (defined at :147) = SparseSoftmaxCrossEntropyWithLogits[T=DT_FLOAT, Tlabels=DT_INT32, _device="/job:localhost/replica:0/task:0/device:GPU:0"](layer_fc_2/BiasAdd, IteratorGetNext/_207)]]
sorry about the post i have figured it out, it was a typo issue that i missed. i created my own tf,record.
but i do get this error
InvalidArgumentError (see above for traceback): logits and labels must have the same first dimension, got logits shape [4,3] and labels shape [1]
[[node SparseSoftmaxCrossEntropyWithLogits/SparseSoftmaxCrossEntropyWithLogits (defined at :147) = SparseSoftmaxCrossEntropyWithLogits[T=DT_FLOAT, Tlabels=DT_INT32, _device="/job:localhost/replica:0/task:0/device:GPU:0"](layer_fc_2/BiasAdd, IteratorGetNext/_207)]]