dadebarr@dadebarr-lambda:~/ptb/ptb-data$ wget https://cross-entropy.net/ML530/ptb-train.py.txt
--2022-11-24 12:12:59--  https://cross-entropy.net/ML530/ptb-train.py.txt
Resolving cross-entropy.net (cross-entropy.net)... 107.180.57.14
Connecting to cross-entropy.net (cross-entropy.net)|107.180.57.14|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1902 (1.9K) [text/plain]
Saving to: ‘ptb-train.py.txt’

ptb-train.py.txt                  100%[============================================================>]   1.86K  --.-KB/s    in 0s

2022-11-24 12:13:00 (543 MB/s) - ‘ptb-train.py.txt’ saved [1902/1902]

dadebarr@dadebarr-lambda:~/ptb/ptb-data$ time python ptb-train.py.txt
2022-11-24 12:13:09.554441: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1532] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 46692 MB memory:  -> device: 0, name: NVIDIA RTX A6000, pci bus id: 0000:21:00.0, compute capability: 8.6
Model: "model"
_________________________________________________________________
 Layer (type)                Output Shape              Param #
=================================================================
 input_1 (InputLayer)        [(None, 250)]             0

 embedding (Embedding)       (None, 250, 1024)         10240000

 lstm (LSTM)                 (None, 1024)              8392704

 tied_embedding (TiedEmbeddi  (None, 10000)            10240000
 ng)

=================================================================
Total params: 18,632,704
Trainable params: 18,632,704
Non-trainable params: 0
_________________________________________________________________
Epoch 1/4
2022-11-24 12:13:12.914291: W tensorflow/core/common_runtime/forward_type_inference.cc:231] Type inference failed. This indicates an invalid graph that escaped type checking. Error message: INVALID_ARGUMENT: expected compatible input types, but input 1:
type_id: TFT_OPTIONAL
args {
  type_id: TFT_PRODUCT
  args {
    type_id: TFT_TENSOR
    args {
      type_id: TFT_LEGACY_VARIANT
    }
  }
}
 is neither a subtype nor a supertype of the combined inputs preceding it:
type_id: TFT_OPTIONAL
args {
  type_id: TFT_PRODUCT
  args {
    type_id: TFT_TENSOR
    args {
      type_id: TFT_FLOAT
    }
  }
}

        while inferring type of node 'cond_40/output/_19'
2022-11-24 12:13:13.903034: I tensorflow/stream_executor/cuda/cuda_dnn.cc:384] Loaded cuDNN version 8303
2022-11-24 12:13:14.078867: I tensorflow/stream_executor/cuda/cuda_blas.cc:1786] TensorFloat-32 will be used for the matrix multiplication. This will only be logged once.
3805/3805 [==============================] - 101s 26ms/step - loss: 5.1778 - sparse_categorical_crossentropy: 4.9132 - val_loss: 4.7897 - val_sparse_categorical_crossentropy: 4.4927
Epoch 2/4
3805/3805 [==============================] - 98s 26ms/step - loss: 4.5709 - sparse_categorical_crossentropy: 4.2374 - val_loss: 4.6354 - val_sparse_categorical_crossentropy: 4.2799
Epoch 3/4
3805/3805 [==============================] - 98s 26ms/step - loss: 4.3390 - sparse_categorical_crossentropy: 3.9486 - val_loss: 4.6286 - val_sparse_categorical_crossentropy: 4.2202
Epoch 4/4
3805/3805 [==============================] - 98s 26ms/step - loss: 4.1802 - sparse_categorical_crossentropy: 3.7391 - val_loss: 4.6613 - val_sparse_categorical_crossentropy: 4.2033
3723/3723 [==============================] - 12s 3ms/step

real    6m54.917s
user    5m58.381s
sys     0m25.810s
dadebarr@dadebarr-lambda:~/ptb/ptb-data$ kaggle competitions submit -c ml530-2022-fall-ptb -f predictions.csv -m "6:55"
100%|██████████████████████████████████████████████████████████████████████████████████████████████| 2.14M/2.14M [00:03<00:00, 682kB/s]
Successfully submitted to ml530-2022-fall-ptbdadebarr@dadebarr-lambda:~/ptb/ptb-data$