-
Hi all,
I'm new to TF and textsum model. I was trying to run the demo by the readme.
When I was trying to train:
```
File "/home/pio/text_sum/ws/textsum/seq2seq_attention_model.py", line 169, …
-
at line number 230
models/textsum/seq2seq_attention_model.py is incorrectly using the api for tf.nn.sampled_softmax_loss,
the order of arg for labels and inputs is reversed in current code
-
In this model,i see the lines in bacth_reader.py:
**while True:
(article, abstract) = input_gen.next()**
i think whether the input_gen…
-
I'm using `node.js` with the `translate` executable that normally would run in pipe like this from the console:
```bash
echo "The quick brown fox jumps over the lazy dog" | ./cli/translate --model…
-
Hi brother
i hava a messy below, when training in BiGRUModel.py :
loss_t = tf.contrib.seq2seq.sequence_loss(
outputs_logits, self.decoder_targets, weig…
-
Can you show ROUGE scores of the model trained on SogouCS corpus? thanks
-
i trained the model using the toy data ,then i decode it the result of decoding is very strange
exapmple :**output= for **
the lastest runing_arg_loss is about 0.002
why cause it? IS the data s…
-
https://github.com/tensorflow/models/blob/master/textsum/seq2seq_attention.py#L163
The way of checking presence of required tokens is wrong because _**WordToId**_ function returns ID of Unknown to…
-
when I excute this instruction:
bazel build -c opt --config=cuda im2txt/...
It throws 'Config values are not defined in any .rc file: cuda', and the training is on CPU mode, not GPU mode
How could I …
-
I use the code "saver.restore(sess, checkpoint_path)" to resume training, but there is a problem when I
observe the loss curve using the tensorboard. It seems that the loss will be reset every some f…