validation/test error on CNNDM
monologue1107 opened this issue · 9 comments
After I trained sum model for CNNDM, I tried to run the validation on my ckpt.
The script I use are followed:
export BERT_DATA_PATH=../bert_data/ export MODEL_PATH=../model/ python train.py \ -task abs \ -mode validate \ -batch_size 3000 \ -test_batch_size 500 \ -bert_data_path ${BERT_DATA_PATH} \ -log_file ../logs/val_abs_bert_cnndm \ -model_path ${MODEL_PATH} \ -sep_optim true \ -use_interval true \ -visible_gpus 1 \ -max_pos 512 \ -max_length 200 \ -alpha 0.95 \ -min_length 50 \ -result_path ../logs/abs_bert_cnndm
When running validation dataset, it went all well. While during test part, it didn't go further by:
`[2020-10-18 20:34:05,003 INFO] Loading valid dataset from ../bert_data/cnndm.valid.6.bert.pt, number of examples: 1362
[2020-10-18 20:34:35,629 INFO] Validation perplexity: 9.66616
[2020-10-18 20:34:35,630 INFO] Validation accuracy: 56.876
[2020-10-18 20:34:35,913 INFO] Loading checkpoint from ../model/model_step_158000.pt
[2020-10-18 20:34:39,414 INFO] loading configuration file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-config.json from cache at ../temp/4dad0251492946e18ac39290fcfe91b89d370fee250efe9521476438fe8ca185.7156163d5fdc189c3016baca0775ffce230789d7fa2a42ef516483e4ca884517
[2020-10-18 20:34:39,415 INFO] Model config {
"architectures": [
"BertForMaskedLM"
],
"attention_probs_dropout_prob": 0.1,
"finetuning_task": null,
"hidden_act": "gelu",
"hidden_dropout_prob": 0.1,
"hidden_size": 768,
"initializer_range": 0.02,
"intermediate_size": 3072,
"layer_norm_eps": 1e-12,
"max_position_embeddings": 512,
"model_type": "bert",
"num_attention_heads": 12,
"num_hidden_layers": 12,
"num_labels": 2,
"output_attentions": false,
"output_hidden_states": false,
"pad_token_id": 0,
"pruned_heads": {},
"torchscript": false,
"type_vocab_size": 2,
"vocab_size": 30522
}
[2020-10-18 20:34:40,460 INFO] loading weights file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-pytorch_model.bin from cache at ../temp/aa1ef1aede4482d0dbcd4d52baad8ae300e60902e88fcb0bebdec09afd232066.36ca03ab34a1a5d5fa7bc3d03d55c4fa650fed07220e2eeebc06ce58d0e9a157
[2020-10-18 20:34:48,701 INFO] Loading test dataset from ../bert_data/cnndm.test.0.bert.pt, number of examples: 2001
[2020-10-18 20:34:49,884 INFO] loading file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt from cache at ../temp/26bc1ad6c0ac742e9b52263248f6d0f00068293b33709fae12320c0e35ccfbbb.542ce4285a40d23a559526243235df47c5f75c197f04f37d1a0c124c32c9a084`
Do you have some opinions about this kind of issue?
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
Have you solved this problem?
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
你好，请问之后你解决这个问题了吗
Have you solved it?
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
Hello, have you solved this problem since then?
你好，请问之后你解决这个问题了吗？
Have you solved it?
Hello, have you solved this problem since then?
你好，请问之后你解决这个问题了吗？
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
Have you solved this problem?
Hello, have you solved this problem since then?
你好，请问之后你解决这个问题了吗？
Yes, I was in the same condition when I try to test my model. Have you solved this problem?
Hello, have you solved this problem since then?
你好，请问之后你解决这个问题了吗？
Have you all solve the problem?