artemfilipenko commited on
Commit
911a3a5
1 Parent(s): be4589a

Upload 13 files

Browse files
README.md ADDED
@@ -0,0 +1,75 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - midas/inspec
7
+ model-index:
8
+ - name: synonyms_5000_plus_3000_default_3_epoch
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # synonyms_5000_plus_3000_default_3_epoch
16
+
17
+ This model is a fine-tuned version of [facebook/bart-large](https://huggingface.co/facebook/bart-large) on the midas/inspec generation dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 1.7956
20
+ - F1@5ext: 0.4590
21
+ - P@5ext: 0.6116
22
+ - R@5ext: 0.4109
23
+ - F1@10ext: 0.5403
24
+ - P@10ext: 0.5953
25
+ - R@10ext: 0.5374
26
+ - F1@5abs: 0.2019
27
+ - P@5abs: 0.3080
28
+ - R@5abs: 0.1721
29
+ - F1@10abs: 0.2307
30
+ - P@10abs: 0.3066
31
+ - R@10abs: 0.2109
32
+ - F1@oext: 0.5427
33
+ - P@oext: 0.6045
34
+ - R@oext: 0.5246
35
+ - F1@oabs: 0.2316
36
+ - P@oabs: 0.3079
37
+ - R@oabs: 0.2094
38
+
39
+ ## Model description
40
+
41
+ More information needed
42
+
43
+ ## Intended uses & limitations
44
+
45
+ More information needed
46
+
47
+ ## Training and evaluation data
48
+
49
+ More information needed
50
+
51
+ ## Training procedure
52
+
53
+ ### Training hyperparameters
54
+
55
+ The following hyperparameters were used during training:
56
+ - learning_rate: 5e-05
57
+ - train_batch_size: 4
58
+ - eval_batch_size: 10
59
+ - seed: 42
60
+ - gradient_accumulation_steps: 6
61
+ - total_train_batch_size: 24
62
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
+ - lr_scheduler_type: linear
64
+ - num_epochs: 3.0
65
+
66
+ ### Training results
67
+
68
+
69
+
70
+ ### Framework versions
71
+
72
+ - Transformers 4.24.0.dev0
73
+ - Pytorch 1.9.0+cu111
74
+ - Datasets 2.6.1
75
+ - Tokenizers 0.13.1
all_results.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_P@10abs": 0.30658571428571446,
4
+ "eval_P@10ext": 0.5953404761904763,
5
+ "eval_P@5abs": 0.30799999999999994,
6
+ "eval_P@5ext": 0.611566666666667,
7
+ "eval_P@Oabs": 0.30785714285714294,
8
+ "eval_P@Oext": 0.6045279220779219,
9
+ "eval_R@10abs": 0.21094011544011554,
10
+ "eval_R@10ext": 0.5373799135848676,
11
+ "eval_R@5abs": 0.17213520923520936,
12
+ "eval_R@5ext": 0.41086155925258133,
13
+ "eval_R@Oabs": 0.2093878066378067,
14
+ "eval_R@Oext": 0.5245630282170534,
15
+ "eval_f1@10abs": 0.23073327724906695,
16
+ "eval_f1@10ext": 0.5403372898595006,
17
+ "eval_f1@5abs": 0.2019063991563992,
18
+ "eval_f1@5ext": 0.45895300009438544,
19
+ "eval_f1@Oabs": 0.2315911764705885,
20
+ "eval_f1@Oext": 0.5427372584935071,
21
+ "eval_loss": 1.7956230640411377,
22
+ "eval_runtime": 325.212,
23
+ "eval_samples": 500,
24
+ "eval_samples_per_second": 1.537,
25
+ "eval_steps_per_second": 0.154,
26
+ "train_loss": 1.0714508148285005,
27
+ "train_runtime": 1254.6583,
28
+ "train_samples": 8000,
29
+ "train_samples_per_second": 19.129,
30
+ "train_steps_per_second": 0.796
31
+ }
config.json ADDED
@@ -0,0 +1,74 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "facebook/bart-large",
3
+ "activation_dropout": 0.1,
4
+ "activation_function": "gelu",
5
+ "add_bias_logits": false,
6
+ "add_final_layer_norm": false,
7
+ "architectures": [
8
+ "BartForConditionalGeneration"
9
+ ],
10
+ "attention_dropout": 0.1,
11
+ "bos_token_id": 0,
12
+ "classif_dropout": 0.1,
13
+ "classifier_dropout": 0.0,
14
+ "d_model": 1024,
15
+ "decoder_attention_heads": 16,
16
+ "decoder_ffn_dim": 4096,
17
+ "decoder_layerdrop": 0.0,
18
+ "decoder_layers": 12,
19
+ "decoder_start_token_id": 2,
20
+ "dropout": 0.1,
21
+ "early_stopping": true,
22
+ "encoder_attention_heads": 16,
23
+ "encoder_ffn_dim": 4096,
24
+ "encoder_layerdrop": 0.0,
25
+ "encoder_layers": 12,
26
+ "eos_token_id": 2,
27
+ "forced_bos_token_id": 0,
28
+ "forced_eos_token_id": 2,
29
+ "gradient_checkpointing": false,
30
+ "id2label": {
31
+ "0": "LABEL_0",
32
+ "1": "LABEL_1",
33
+ "2": "LABEL_2"
34
+ },
35
+ "init_std": 0.02,
36
+ "is_encoder_decoder": true,
37
+ "label2id": {
38
+ "LABEL_0": 0,
39
+ "LABEL_1": 1,
40
+ "LABEL_2": 2
41
+ },
42
+ "max_position_embeddings": 1024,
43
+ "model_type": "bart",
44
+ "no_repeat_ngram_size": 3,
45
+ "normalize_before": false,
46
+ "num_beams": 20,
47
+ "num_hidden_layers": 12,
48
+ "pad_token_id": 1,
49
+ "scale_embedding": false,
50
+ "task_specific_params": {
51
+ "summarization": {
52
+ "length_penalty": 1.0,
53
+ "max_length": 128,
54
+ "min_length": 12,
55
+ "num_beams": 4
56
+ },
57
+ "summarization_cnn": {
58
+ "length_penalty": 2.0,
59
+ "max_length": 142,
60
+ "min_length": 56,
61
+ "num_beams": 4
62
+ },
63
+ "summarization_xsum": {
64
+ "length_penalty": 1.0,
65
+ "max_length": 62,
66
+ "min_length": 11,
67
+ "num_beams": 6
68
+ }
69
+ },
70
+ "torch_dtype": "float32",
71
+ "transformers_version": "4.24.0.dev0",
72
+ "use_cache": true,
73
+ "vocab_size": 50265
74
+ }
eval_results.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "eval_P@10abs": 0.30658571428571446,
4
+ "eval_P@10ext": 0.5953404761904763,
5
+ "eval_P@5abs": 0.30799999999999994,
6
+ "eval_P@5ext": 0.611566666666667,
7
+ "eval_P@Oabs": 0.30785714285714294,
8
+ "eval_P@Oext": 0.6045279220779219,
9
+ "eval_R@10abs": 0.21094011544011554,
10
+ "eval_R@10ext": 0.5373799135848676,
11
+ "eval_R@5abs": 0.17213520923520936,
12
+ "eval_R@5ext": 0.41086155925258133,
13
+ "eval_R@Oabs": 0.2093878066378067,
14
+ "eval_R@Oext": 0.5245630282170534,
15
+ "eval_f1@10abs": 0.23073327724906695,
16
+ "eval_f1@10ext": 0.5403372898595006,
17
+ "eval_f1@5abs": 0.2019063991563992,
18
+ "eval_f1@5ext": 0.45895300009438544,
19
+ "eval_f1@Oabs": 0.2315911764705885,
20
+ "eval_f1@Oext": 0.5427372584935071,
21
+ "eval_loss": 1.7956230640411377,
22
+ "eval_runtime": 325.212,
23
+ "eval_samples": 500,
24
+ "eval_samples_per_second": 1.537,
25
+ "eval_steps_per_second": 0.154
26
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f6621168733888eb9445a2d10e32b5dcf04999fbfa29103011758aba70fdd30f
3
+ size 1625537793
special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<s>",
4
+ "cls_token": "<s>",
5
+ "eos_token": "</s>",
6
+ "errors": "replace",
7
+ "mask_token": "<mask>",
8
+ "model_max_length": 1024,
9
+ "name_or_path": "facebook/bart-large",
10
+ "pad_token": "<pad>",
11
+ "sep_token": "</s>",
12
+ "special_tokens_map_file": null,
13
+ "tokenizer_class": "BartTokenizer",
14
+ "trim_offsets": true,
15
+ "unk_token": "<unk>"
16
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 1.0714508148285005,
4
+ "train_runtime": 1254.6583,
5
+ "train_samples": 8000,
6
+ "train_samples_per_second": 19.129,
7
+ "train_steps_per_second": 0.796
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.999,
5
+ "global_step": 999,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 1.5,
12
+ "learning_rate": 2.4974974974974976e-05,
13
+ "loss": 1.3409,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 3.0,
18
+ "step": 999,
19
+ "total_flos": 1.4606682885095424e+16,
20
+ "train_loss": 1.0714508148285005,
21
+ "train_runtime": 1254.6583,
22
+ "train_samples_per_second": 19.129,
23
+ "train_steps_per_second": 0.796
24
+ }
25
+ ],
26
+ "max_steps": 999,
27
+ "num_train_epochs": 3,
28
+ "total_flos": 1.4606682885095424e+16,
29
+ "trial_name": null,
30
+ "trial_params": null
31
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eddb7b862bea04a1c57e11e923e57d60c15c880bfbf75db6e06f724ac4f68e07
3
+ size 3567
vocab.json ADDED
The diff for this file is too large to render. See raw diff