| 2020-09-19 12:19:01,737 loading file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt from cache at /home/tiger/.cache/torch/transformers/26bc1ad6c0ac742e9b52263248f6d0f00068293b33709fae12320c0e35ccfbbb.542ce4285a40d23a559526243235df47c5f75c197f04f37d1a0c124c32c9a084 |
| 2020-09-19 12:19:03,261 loading configuration file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-config.json from cache at /home/tiger/.cache/torch/transformers/4dad0251492946e18ac39290fcfe91b89d370fee250efe9521476438fe8ca185.7156163d5fdc189c3016baca0775ffce230789d7fa2a42ef516483e4ca884517 |
| 2020-09-19 12:19:03,261 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, |
| "use_bfloat16": false, |
| "vocab_size": 30522 |
| } |
|
|
| 2020-09-19 12:19:04,756 loading weights file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-pytorch_model.bin from cache at /home/tiger/.cache/torch/transformers/aa1ef1aede4482d0dbcd4d52baad8ae300e60902e88fcb0bebdec09afd232066.36ca03ab34a1a5d5fa7bc3d03d55c4fa650fed07220e2eeebc06ce58d0e9a157 |
| 2020-09-19 12:19:07,026 Use adam as the optimizer, with init lr 0.0005 |
| 2020-09-19 12:19:07,026 Image encoder is data paralleled now. |
| 2020-09-19 12:19:07,039 Loading dataset |
| 2020-09-19 12:21:15,606 Computing results... |
| 2020-09-19 12:21:32,949 Test: [0/196] Le 62.1216 (62.1216) Time 17.342 (0.000) |
| 2020-09-19 12:21:34,949 Test: [10/196] Le 62.3208 (62.0296) Time 0.212 (0.000) |
| 2020-09-19 12:21:36,924 Test: [20/196] Le 61.6259 (61.7868) Time 0.196 (0.000) |
| 2020-09-19 12:21:38,923 Test: [30/196] Le 62.5433 (61.8626) Time 0.197 (0.000) |
| 2020-09-19 12:21:40,909 Test: [40/196] Le 62.5389 (61.6597) Time 0.192 (0.000) |
| 2020-09-19 12:21:42,868 Test: [50/196] Le 62.0505 (61.6832) Time 0.204 (0.000) |
| 2020-09-19 12:21:44,851 Test: [60/196] Le 60.7455 (61.7560) Time 0.199 (0.000) |
| 2020-09-19 12:21:46,795 Test: [70/196] Le 63.2594 (61.6683) Time 0.194 (0.000) |
| 2020-09-19 12:21:48,835 Test: [80/196] Le 62.5849 (61.7312) Time 0.220 (0.000) |
| 2020-09-19 12:21:50,822 Test: [90/196] Le 61.6402 (61.7597) Time 0.196 (0.000) |
| 2020-09-19 12:21:52,830 Test: [100/196] Le 63.8944 (61.8340) Time 0.198 (0.000) |
| 2020-09-19 12:21:54,811 Test: [110/196] Le 61.2093 (61.8118) Time 0.190 (0.000) |
| 2020-09-19 12:21:56,822 Test: [120/196] Le 66.6486 (61.8693) Time 0.203 (0.000) |
| 2020-09-19 12:21:58,862 Test: [130/196] Le 60.1235 (61.8624) Time 0.202 (0.000) |
| 2020-09-19 12:22:00,809 Test: [140/196] Le 64.4691 (61.9337) Time 0.188 (0.000) |
| 2020-09-19 12:22:02,771 Test: [150/196] Le 59.6040 (61.9000) Time 0.193 (0.000) |
| 2020-09-19 12:22:04,715 Test: [160/196] Le 62.9591 (61.9137) Time 0.191 (0.000) |
| 2020-09-19 12:22:06,701 Test: [170/196] Le 61.7752 (61.8860) Time 0.200 (0.000) |
| 2020-09-19 12:22:08,728 Test: [180/196] Le 60.3819 (61.8454) Time 0.193 (0.000) |
| 2020-09-19 12:22:10,707 Test: [190/196] Le 59.6309 (61.8942) Time 0.193 (0.000) |
| 2020-09-19 12:22:11,655 Images: 5000, Captions: 25000 |
| 2020-09-19 12:22:45,146 calculate similarity time: |
| 2020-09-19 12:22:45,524 Image to text: 81.9, 97.3, 99.2, 1.0, 1.5 |
| 2020-09-19 12:22:45,844 Text to image: 66.7, 92.0, 96.3, 1.0, 4.2 |
| 2020-09-19 12:22:45,845 rsum: 533.4 ar: 92.8 ari: 85.0 |
| 2020-09-19 12:22:45,947 calculate similarity time: |
| 2020-09-19 12:22:46,351 Image to text: 79.0, 96.2, 98.5, 1.0, 1.8 |
| 2020-09-19 12:22:46,672 Text to image: 64.7, 90.6, 95.8, 1.0, 4.2 |
| 2020-09-19 12:22:46,672 rsum: 524.8 ar: 91.2 ari: 83.7 |
| 2020-09-19 12:22:46,753 calculate similarity time: |
| 2020-09-19 12:22:47,156 Image to text: 79.6, 96.2, 98.8, 1.0, 1.6 |
| 2020-09-19 12:22:47,476 Text to image: 64.4, 91.2, 96.1, 1.0, 3.7 |
| 2020-09-19 12:22:47,476 rsum: 526.3 ar: 91.5 ari: 83.9 |
| 2020-09-19 12:22:47,568 calculate similarity time: |
| 2020-09-19 12:22:47,974 Image to text: 79.3, 96.1, 98.8, 1.0, 1.7 |
| 2020-09-19 12:22:48,295 Text to image: 62.4, 91.1, 96.4, 1.0, 3.3 |
| 2020-09-19 12:22:48,296 rsum: 524.2 ar: 91.4 ari: 83.3 |
| 2020-09-19 12:22:48,375 calculate similarity time: |
| 2020-09-19 12:22:48,778 Image to text: 78.5, 96.2, 99.1, 1.0, 1.6 |
| 2020-09-19 12:22:49,099 Text to image: 65.7, 92.2, 96.8, 1.0, 3.6 |
| 2020-09-19 12:22:49,099 rsum: 528.6 ar: 91.3 ari: 84.9 |
| 2020-09-19 12:22:49,099 ----------------------------------- |
| 2020-09-19 12:22:49,099 Mean metrics: |
| 2020-09-19 12:22:49,099 rsum: 527.5 |
| 2020-09-19 12:22:49,099 Average i2t Recall: 91.6 |
| 2020-09-19 12:22:49,099 Image to text: 79.7 96.4 98.9 1.0 1.6 |
| 2020-09-19 12:22:49,099 Average t2i Recall: 84.2 |
| 2020-09-19 12:22:49,099 Text to image: 64.8 91.4 96.3 1.0 3.8 |
|
|
|
|
| 2020-09-19 12:22:51,067 loading file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-vocab.txt from cache at /home/tiger/.cache/torch/transformers/26bc1ad6c0ac742e9b52263248f6d0f00068293b33709fae12320c0e35ccfbbb.542ce4285a40d23a559526243235df47c5f75c197f04f37d1a0c124c32c9a084 |
| 2020-09-19 12:22:52,532 loading configuration file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-config.json from cache at /home/tiger/.cache/torch/transformers/4dad0251492946e18ac39290fcfe91b89d370fee250efe9521476438fe8ca185.7156163d5fdc189c3016baca0775ffce230789d7fa2a42ef516483e4ca884517 |
| 2020-09-19 12:22:52,533 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, |
| "use_bfloat16": false, |
| "vocab_size": 30522 |
| } |
|
|
| 2020-09-19 12:22:53,926 loading weights file https://s3.amazonaws.com/models.huggingface.co/bert/bert-base-uncased-pytorch_model.bin from cache at /home/tiger/.cache/torch/transformers/aa1ef1aede4482d0dbcd4d52baad8ae300e60902e88fcb0bebdec09afd232066.36ca03ab34a1a5d5fa7bc3d03d55c4fa650fed07220e2eeebc06ce58d0e9a157 |
| 2020-09-19 12:22:56,137 Use adam as the optimizer, with init lr 0.0005 |
| 2020-09-19 12:22:56,137 Image encoder is data paralleled now. |
| 2020-09-19 12:22:56,150 Loading dataset |
| 2020-09-19 12:25:11,707 Computing results... |
| 2020-09-19 12:25:11,921 Test: [0/196] Le 62.1216 (62.1216) Time 0.212 (0.000) |
| 2020-09-19 12:25:13,893 Test: [10/196] Le 62.3208 (62.0296) Time 0.210 (0.000) |
| 2020-09-19 12:25:15,880 Test: [20/196] Le 61.6259 (61.7868) Time 0.197 (0.000) |
| 2020-09-19 12:25:17,889 Test: [30/196] Le 62.5433 (61.8626) Time 0.230 (0.000) |
| 2020-09-19 12:25:19,883 Test: [40/196] Le 62.5389 (61.6597) Time 0.192 (0.000) |
| 2020-09-19 12:25:21,849 Test: [50/196] Le 62.0505 (61.6832) Time 0.198 (0.000) |
| 2020-09-19 12:25:23,844 Test: [60/196] Le 60.7455 (61.7560) Time 0.232 (0.000) |
| 2020-09-19 12:25:25,804 Test: [70/196] Le 63.2594 (61.6683) Time 0.197 (0.000) |
| 2020-09-19 12:25:27,861 Test: [80/196] Le 62.5849 (61.7312) Time 0.220 (0.000) |
| 2020-09-19 12:25:29,850 Test: [90/196] Le 61.6402 (61.7597) Time 0.201 (0.000) |
| 2020-09-19 12:25:31,876 Test: [100/196] Le 63.8944 (61.8340) Time 0.201 (0.000) |
| 2020-09-19 12:25:33,838 Test: [110/196] Le 61.2093 (61.8118) Time 0.191 (0.000) |
| 2020-09-19 12:25:35,804 Test: [120/196] Le 66.6486 (61.8693) Time 0.202 (0.000) |
| 2020-09-19 12:25:37,812 Test: [130/196] Le 60.1235 (61.8624) Time 0.200 (0.000) |
| 2020-09-19 12:25:39,779 Test: [140/196] Le 64.4691 (61.9337) Time 0.192 (0.000) |
| 2020-09-19 12:25:41,746 Test: [150/196] Le 59.6040 (61.9000) Time 0.198 (0.000) |
| 2020-09-19 12:25:43,717 Test: [160/196] Le 62.9591 (61.9137) Time 0.197 (0.000) |
| 2020-09-19 12:25:45,698 Test: [170/196] Le 61.7752 (61.8860) Time 0.203 (0.000) |
| 2020-09-19 12:25:47,779 Test: [180/196] Le 60.3819 (61.8454) Time 0.198 (0.000) |
| 2020-09-19 12:25:49,797 Test: [190/196] Le 59.6309 (61.8942) Time 0.194 (0.000) |
| 2020-09-19 12:25:50,722 Images: 5000, Captions: 25000 |
| 2020-09-19 12:26:45,893 rsum: 434.3 |
| 2020-09-19 12:26:45,894 Average i2t Recall: 78.7 |
| 2020-09-19 12:26:45,894 Image to text: 58.3 85.3 92.3 1.0 4.1 |
| 2020-09-19 12:26:45,894 Average t2i Recall: 66.1 |
| 2020-09-19 12:26:45,894 Text to image: 42.4 72.7 83.2 2.0 14.8 |
|
|
