janbakker commited on
Commit
0763f3c
1 Parent(s): fdafce2

Upload LEDForConditionalGeneration

Browse files
Files changed (3) hide show
  1. config.json +155 -0
  2. generation_config.json +8 -0
  3. pytorch_model.bin +3 -0
config.json ADDED
@@ -0,0 +1,155 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "allenai/led-base-16384",
3
+ "accelerator": null,
4
+ "accumulate_grad_batches": 2,
5
+ "activation_dropout": 0.0,
6
+ "activation_function": "gelu",
7
+ "add_context": false,
8
+ "amp_backend": null,
9
+ "amp_level": null,
10
+ "architectures": [
11
+ "LEDForConditionalGeneration"
12
+ ],
13
+ "attention_dropout": 0.0,
14
+ "attention_window": [
15
+ 1024,
16
+ 1024,
17
+ 1024,
18
+ 1024,
19
+ 1024,
20
+ 1024
21
+ ],
22
+ "auto_lr_find": false,
23
+ "auto_scale_batch_size": false,
24
+ "auto_select_gpus": null,
25
+ "batch_size": 8,
26
+ "benchmark": null,
27
+ "bos_token_id": 0,
28
+ "check_val_every_n_epoch": 1,
29
+ "checkpoint": null,
30
+ "class_labels": {
31
+ "delete": 3,
32
+ "ignore": 0,
33
+ "rephrase": 1,
34
+ "split": 2
35
+ },
36
+ "classif_dropout": 0.0,
37
+ "classifier_dropout": 0.0,
38
+ "context_dir": null,
39
+ "context_doc_id": null,
40
+ "context_window": 13,
41
+ "d_model": 768,
42
+ "decoder_attention_heads": 12,
43
+ "decoder_ffn_dim": 3072,
44
+ "decoder_layerdrop": 0.0,
45
+ "decoder_layers": 6,
46
+ "decoder_start_token_id": 2,
47
+ "default_root_dir": null,
48
+ "detect_anomaly": false,
49
+ "devices": "2",
50
+ "dropout": 0.1,
51
+ "enable_checkpointing": true,
52
+ "enable_model_summary": true,
53
+ "enable_progress_bar": true,
54
+ "encoder_attention_heads": 12,
55
+ "encoder_ffn_dim": 3072,
56
+ "encoder_layerdrop": 0.0,
57
+ "encoder_layers": 6,
58
+ "eos_token_id": 2,
59
+ "eval_beams": null,
60
+ "eval_max_length": null,
61
+ "fast_dev_run": false,
62
+ "gpus": null,
63
+ "gradient_checkpointing": false,
64
+ "gradient_clip_algorithm": null,
65
+ "gradient_clip_val": null,
66
+ "id2label": {
67
+ "0": "LABEL_0",
68
+ "1": "LABEL_1",
69
+ "2": "LABEL_2"
70
+ },
71
+ "inference_mode": true,
72
+ "init_std": 0.02,
73
+ "ipus": null,
74
+ "is_encoder_decoder": true,
75
+ "label2id": {
76
+ "LABEL_0": 0,
77
+ "LABEL_1": 1,
78
+ "LABEL_2": 2
79
+ },
80
+ "limit_predict_batches": null,
81
+ "limit_test_batches": null,
82
+ "limit_train_batches": null,
83
+ "limit_val_batches": null,
84
+ "log_every_n_steps": 50,
85
+ "logger": true,
86
+ "longformer": true,
87
+ "lr": 2e-05,
88
+ "lr_scheduler": false,
89
+ "max_decoder_position_embeddings": 1024,
90
+ "max_encoder_position_embeddings": 16384,
91
+ "max_epochs": null,
92
+ "max_length": 1024,
93
+ "max_samples": -1,
94
+ "max_steps": -1,
95
+ "max_time": null,
96
+ "min_epochs": null,
97
+ "min_steps": null,
98
+ "model_type": "led",
99
+ "move_metrics_to_cpu": false,
100
+ "multiple_trainloader_mode": "max_size_cycle",
101
+ "name": "o-led-para",
102
+ "num_hidden_layers": 6,
103
+ "num_nodes": 1,
104
+ "num_processes": null,
105
+ "num_sanity_val_steps": 2,
106
+ "op_col": "label",
107
+ "op_tokens": [
108
+ "<COPY>",
109
+ "<REPHRASE>",
110
+ "<SPLIT>",
111
+ "<DELETE>"
112
+ ],
113
+ "overfit_batches": 0.0,
114
+ "pad_token_id": 1,
115
+ "plan_col": "labels",
116
+ "plan_loss": null,
117
+ "plan_prefix": false,
118
+ "plan_sep": false,
119
+ "plugins": null,
120
+ "precision": 32,
121
+ "prefix_only": false,
122
+ "profiler": null,
123
+ "project": "simplification_models",
124
+ "reading_lvl": null,
125
+ "reload_dataloaders_every_n_epochs": 0,
126
+ "replace_sampler_ddp": true,
127
+ "resume_from_checkpoint": null,
128
+ "save_dir": null,
129
+ "sent_level": true,
130
+ "simple_context_dir": null,
131
+ "simple_context_doc_id": null,
132
+ "skip_val_gen": true,
133
+ "strategy": null,
134
+ "sync_batchnorm": false,
135
+ "torch_dtype": "float32",
136
+ "tpu_cores": null,
137
+ "track_grad_norm": -1,
138
+ "train_check_interval": 0.01,
139
+ "train_data_dir": null,
140
+ "train_file": "data/wikiauto_para_train.csv",
141
+ "train_split": 0.9,
142
+ "train_workers": 32,
143
+ "transformers_version": "4.29.1",
144
+ "use_cache": true,
145
+ "val_check_interval": null,
146
+ "val_file": "data/wikiauto_para_valid.csv",
147
+ "val_metric": "bleu",
148
+ "val_split": 0.05,
149
+ "val_workers": 8,
150
+ "valid_data_dir": null,
151
+ "vocab_size": 50274,
152
+ "wandb_id": null,
153
+ "x_col": "complex",
154
+ "y_col": "simple"
155
+ }
generation_config.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "decoder_start_token_id": 2,
5
+ "eos_token_id": 2,
6
+ "pad_token_id": 1,
7
+ "transformers_version": "4.29.1"
8
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:603b4b03d583d650a49bfab42e554d70e3b7a616d8f73a5a46d83b19a679a62d
3
+ size 647708525