Hugging Face's logo

ml6team
/
keyphrase-generation-t5-small-inspec

Text2Text Generation
Transformers
PyTorch
English
t5
keyphrase-generation
Eval Results
text-generation-inference
Inference Endpoints
Model card Files Community
3
Thomas De Decker commited on
Commit
2b941d3
•
1 Parent(s): e8044d4

Add model

Browse files
Files changed (13) hide show
  1. added_tokens.json +1 -0
  2. config.json +57 -0
  3. optimizer.pt +3 -0
  4. pytorch_model.bin +3 -0
  5. rng_state.pth +3 -0
  6. scaler.pt +3 -0
  7. scheduler.pt +3 -0
  8. special_tokens_map.json +1 -0
  9. spiece.model +3 -0
  10. tokenizer.json +0 -0
  11. tokenizer_config.json +1 -0
  12. trainer_state.json +976 -0
  13. training_args.bin +3 -0
added_tokens.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"[KP_SEP]": 32100}
config.json ADDED
@@ -0,0 +1,57 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "t5-small",
3
+ "architectures": [
4
+ "T5ForConditionalGeneration"
5
+ ],
6
+ "d_ff": 2048,
7
+ "d_kv": 64,
8
+ "d_model": 512,
9
+ "decoder_start_token_id": 0,
10
+ "dropout_rate": 0.1,
11
+ "eos_token_id": 1,
12
+ "feed_forward_proj": "relu",
13
+ "initializer_factor": 1.0,
14
+ "is_encoder_decoder": true,
15
+ "layer_norm_epsilon": 1e-06,
16
+ "model_type": "t5",
17
+ "n_positions": 512,
18
+ "num_decoder_layers": 6,
19
+ "num_heads": 8,
20
+ "num_layers": 6,
21
+ "output_past": true,
22
+ "pad_token_id": 0,
23
+ "relative_attention_num_buckets": 32,
24
+ "task_specific_params": {
25
+ "summarization": {
26
+ "early_stopping": true,
27
+ "length_penalty": 2.0,
28
+ "max_length": 200,
29
+ "min_length": 30,
30
+ "no_repeat_ngram_size": 3,
31
+ "num_beams": 4,
32
+ "prefix": "summarize: "
33
+ },
34
+ "translation_en_to_de": {
35
+ "early_stopping": true,
36
+ "max_length": 300,
37
+ "num_beams": 4,
38
+ "prefix": "translate English to German: "
39
+ },
40
+ "translation_en_to_fr": {
41
+ "early_stopping": true,
42
+ "max_length": 300,
43
+ "num_beams": 4,
44
+ "prefix": "translate English to French: "
45
+ },
46
+ "translation_en_to_ro": {
47
+ "early_stopping": true,
48
+ "max_length": 300,
49
+ "num_beams": 4,
50
+ "prefix": "translate English to Romanian: "
51
+ }
52
+ },
53
+ "torch_dtype": "float32",
54
+ "transformers_version": "4.17.0",
55
+ "use_cache": true,
56
+ "vocab_size": 32101
57
+ }
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:749043cdff035d1ab2240dc75e407bd10596fb9bf80e36333404fcc387cc207d
3
+ size 484018869
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:67394bed029cb1b3f08c8acf4c65f7cb03b273e38af19fc879982de6f233e2e6
3
+ size 242030331
rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:36b59e660513c24c68e47eed7aa58cf2063c3481b9670625d35ade3d2691c30d
3
+ size 14503
scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f460c9860f92d63723a55e876b98b52e4a8c8c8716d122bba2e0436a80cea0be
3
+ size 559
scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e8cf943a69f9855f9d67f54e487e99ae49506ca61750728a12a0aa7d6c984206
3
+ size 623
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"]}
spiece.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d60acb128cf7b7f2536e8f38a5b18a05535c9e14c7a355904270e15b0945ea86
3
+ size 791656
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"eos_token": "</s>", "unk_token": "<unk>", "pad_token": "<pad>", "extra_ids": 100, "additional_special_tokens": ["<extra_id_0>", "<extra_id_1>", "<extra_id_2>", "<extra_id_3>", "<extra_id_4>", "<extra_id_5>", "<extra_id_6>", "<extra_id_7>", "<extra_id_8>", "<extra_id_9>", "<extra_id_10>", "<extra_id_11>", "<extra_id_12>", "<extra_id_13>", "<extra_id_14>", "<extra_id_15>", "<extra_id_16>", "<extra_id_17>", "<extra_id_18>", "<extra_id_19>", "<extra_id_20>", "<extra_id_21>", "<extra_id_22>", "<extra_id_23>", "<extra_id_24>", "<extra_id_25>", "<extra_id_26>", "<extra_id_27>", "<extra_id_28>", "<extra_id_29>", "<extra_id_30>", "<extra_id_31>", "<extra_id_32>", "<extra_id_33>", "<extra_id_34>", "<extra_id_35>", "<extra_id_36>", "<extra_id_37>", "<extra_id_38>", "<extra_id_39>", "<extra_id_40>", "<extra_id_41>", "<extra_id_42>", "<extra_id_43>", "<extra_id_44>", "<extra_id_45>", "<extra_id_46>", "<extra_id_47>", "<extra_id_48>", "<extra_id_49>", "<extra_id_50>", "<extra_id_51>", "<extra_id_52>", "<extra_id_53>", "<extra_id_54>", "<extra_id_55>", "<extra_id_56>", "<extra_id_57>", "<extra_id_58>", "<extra_id_59>", "<extra_id_60>", "<extra_id_61>", "<extra_id_62>", "<extra_id_63>", "<extra_id_64>", "<extra_id_65>", "<extra_id_66>", "<extra_id_67>", "<extra_id_68>", "<extra_id_69>", "<extra_id_70>", "<extra_id_71>", "<extra_id_72>", "<extra_id_73>", "<extra_id_74>", "<extra_id_75>", "<extra_id_76>", "<extra_id_77>", "<extra_id_78>", "<extra_id_79>", "<extra_id_80>", "<extra_id_81>", "<extra_id_82>", "<extra_id_83>", "<extra_id_84>", "<extra_id_85>", "<extra_id_86>", "<extra_id_87>", "<extra_id_88>", "<extra_id_89>", "<extra_id_90>", "<extra_id_91>", "<extra_id_92>", "<extra_id_93>", "<extra_id_94>", "<extra_id_95>", "<extra_id_96>", "<extra_id_97>", "<extra_id_98>", "<extra_id_99>"], "model_max_length": 512, "special_tokens_map_file": null, "name_or_path": "t5-small", "tokenizer_class": "T5Tokenizer"}
trainer_state.json ADDED
@@ -0,0 +1,976 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.3611119221851426,
3
+ "best_model_checkpoint": "t5-small-keyword-generation-inspec\\checkpoint-20000",
4
+ "epoch": 20.0,
5
+ "global_step": 20000,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.5,
12
+ "learning_rate": 4.93e-05,
13
+ "loss": 7.9531,
14
+ "step": 500
15
+ },
16
+ {
17
+ "epoch": 0.5,
18
+ "eval_F1@10": 0.0,
19
+ "eval_F1@5": 0.0,
20
+ "eval_F1@M": 0.0,
21
+ "eval_P@10": 0.0,
22
+ "eval_P@5": 0.0,
23
+ "eval_P@M": 0.0,
24
+ "eval_R@10": 0.0,
25
+ "eval_R@5": 0.0,
26
+ "eval_R@M": 0.0,
27
+ "eval_loss": 4.233538627624512,
28
+ "eval_model_name": "t5-small-keyword-generation-inspec",
29
+ "eval_runtime": 294.894,
30
+ "eval_samples_per_second": 1.696,
31
+ "eval_steps_per_second": 1.696,
32
+ "step": 500
33
+ },
34
+ {
35
+ "epoch": 1.0,
36
+ "learning_rate": 9.93e-05,
37
+ "loss": 3.9646,
38
+ "step": 1000
39
+ },
40
+ {
41
+ "epoch": 1.0,
42
+ "eval_F1@10": 0.1977849653076471,
43
+ "eval_F1@5": 0.2771428832605307,
44
+ "eval_F1@M": 0.3029857823291265,
45
+ "eval_P@10": 0.17059999999999995,
46
+ "eval_P@5": 0.33559999999999995,
47
+ "eval_P@M": 0.42588095238095236,
48
+ "eval_R@10": 0.2556045288045292,
49
+ "eval_R@5": 0.251349622599623,
50
+ "eval_R@M": 0.2556045288045292,
51
+ "eval_loss": 2.812638282775879,
52
+ "eval_model_name": "t5-small-keyword-generation-inspec",
53
+ "eval_runtime": 298.6756,
54
+ "eval_samples_per_second": 1.674,
55
+ "eval_steps_per_second": 1.674,
56
+ "step": 1000
57
+ },
58
+ {
59
+ "epoch": 1.5,
60
+ "learning_rate": 9.899387755102041e-05,
61
+ "loss": 2.9718,
62
+ "step": 1500
63
+ },
64
+ {
65
+ "epoch": 1.5,
66
+ "eval_F1@10": 0.21689912919673357,
67
+ "eval_F1@5": 0.303470865735572,
68
+ "eval_F1@M": 0.3314612888315681,
69
+ "eval_P@10": 0.18799999999999992,
70
+ "eval_P@5": 0.3703999999999999,
71
+ "eval_P@M": 0.4614,
72
+ "eval_R@10": 0.2789250194250196,
73
+ "eval_R@5": 0.27387566877566905,
74
+ "eval_R@M": 0.2789250194250196,
75
+ "eval_loss": 2.4728078842163086,
76
+ "eval_model_name": "t5-small-keyword-generation-inspec",
77
+ "eval_runtime": 307.3423,
78
+ "eval_samples_per_second": 1.627,
79
+ "eval_steps_per_second": 1.627,
80
+ "step": 1500
81
+ },
82
+ {
83
+ "epoch": 2.0,
84
+ "learning_rate": 9.79734693877551e-05,
85
+ "loss": 2.5763,
86
+ "step": 2000
87
+ },
88
+ {
89
+ "epoch": 2.0,
90
+ "eval_F1@10": 0.21342376898165072,
91
+ "eval_F1@5": 0.29715987117457776,
92
+ "eval_F1@M": 0.3246857584795672,
93
+ "eval_P@10": 0.18480000000000008,
94
+ "eval_P@5": 0.3628000000000001,
95
+ "eval_P@M": 0.44907619047619074,
96
+ "eval_R@10": 0.2744417082917087,
97
+ "eval_R@5": 0.2677502941502946,
98
+ "eval_R@M": 0.2744417082917087,
99
+ "eval_loss": 2.183018445968628,
100
+ "eval_model_name": "t5-small-keyword-generation-inspec",
101
+ "eval_runtime": 289.0824,
102
+ "eval_samples_per_second": 1.73,
103
+ "eval_steps_per_second": 1.73,
104
+ "step": 2000
105
+ },
106
+ {
107
+ "epoch": 2.5,
108
+ "learning_rate": 9.695510204081633e-05,
109
+ "loss": 2.2131,
110
+ "step": 2500
111
+ },
112
+ {
113
+ "epoch": 2.5,
114
+ "eval_F1@10": 0.22083437505943945,
115
+ "eval_F1@5": 0.3085772965596502,
116
+ "eval_F1@M": 0.33408261078446894,
117
+ "eval_P@10": 0.1906,
118
+ "eval_P@5": 0.3736000000000001,
119
+ "eval_P@M": 0.45226190476190464,
120
+ "eval_R@10": 0.2875526029526035,
121
+ "eval_R@5": 0.28179986124986167,
122
+ "eval_R@M": 0.2875526029526035,
123
+ "eval_loss": 2.037614107131958,
124
+ "eval_model_name": "t5-small-keyword-generation-inspec",
125
+ "eval_runtime": 294.2365,
126
+ "eval_samples_per_second": 1.699,
127
+ "eval_steps_per_second": 1.699,
128
+ "step": 2500
129
+ },
130
+ {
131
+ "epoch": 3.0,
132
+ "learning_rate": 9.593469387755103e-05,
133
+ "loss": 2.0094,
134
+ "step": 3000
135
+ },
136
+ {
137
+ "epoch": 3.0,
138
+ "eval_F1@10": 0.23622480848222596,
139
+ "eval_F1@5": 0.32680995997466633,
140
+ "eval_F1@M": 0.3509576027172007,
141
+ "eval_P@10": 0.2044,
142
+ "eval_P@5": 0.39799999999999996,
143
+ "eval_P@M": 0.4652142857142857,
144
+ "eval_R@10": 0.30521768231768254,
145
+ "eval_R@5": 0.29608175158175193,
146
+ "eval_R@M": 0.30521768231768254,
147
+ "eval_loss": 1.9597641229629517,
148
+ "eval_model_name": "t5-small-keyword-generation-inspec",
149
+ "eval_runtime": 299.8378,
150
+ "eval_samples_per_second": 1.668,
151
+ "eval_steps_per_second": 1.668,
152
+ "step": 3000
153
+ },
154
+ {
155
+ "epoch": 3.5,
156
+ "learning_rate": 9.491428571428572e-05,
157
+ "loss": 1.7087,
158
+ "step": 3500
159
+ },
160
+ {
161
+ "epoch": 3.5,
162
+ "eval_F1@10": 0.23585947180128827,
163
+ "eval_F1@5": 0.32583185474656123,
164
+ "eval_F1@M": 0.3484568159051137,
165
+ "eval_P@10": 0.20420000000000005,
166
+ "eval_P@5": 0.39679999999999993,
167
+ "eval_P@M": 0.457257142857143,
168
+ "eval_R@10": 0.3021705960705962,
169
+ "eval_R@5": 0.2934577533577537,
170
+ "eval_R@M": 0.3021705960705962,
171
+ "eval_loss": 1.9531850814819336,
172
+ "eval_model_name": "t5-small-keyword-generation-inspec",
173
+ "eval_runtime": 291.44,
174
+ "eval_samples_per_second": 1.716,
175
+ "eval_steps_per_second": 1.716,
176
+ "step": 3500
177
+ },
178
+ {
179
+ "epoch": 4.0,
180
+ "learning_rate": 9.389387755102042e-05,
181
+ "loss": 1.7346,
182
+ "step": 4000
183
+ },
184
+ {
185
+ "epoch": 4.0,
186
+ "eval_F1@10": 0.23949844293171174,
187
+ "eval_F1@5": 0.3324181589652182,
188
+ "eval_F1@M": 0.3555839600200287,
189
+ "eval_P@10": 0.20739999999999992,
190
+ "eval_P@5": 0.40439999999999987,
191
+ "eval_P@M": 0.46959285714285726,
192
+ "eval_R@10": 0.307689471639472,
193
+ "eval_R@5": 0.30029141969142004,
194
+ "eval_R@M": 0.307689471639472,
195
+ "eval_loss": 1.8844341039657593,
196
+ "eval_model_name": "t5-small-keyword-generation-inspec",
197
+ "eval_runtime": 279.8232,
198
+ "eval_samples_per_second": 1.787,
199
+ "eval_steps_per_second": 1.787,
200
+ "step": 4000
201
+ },
202
+ {
203
+ "epoch": 4.5,
204
+ "learning_rate": 9.287551020408164e-05,
205
+ "loss": 1.4093,
206
+ "step": 4500
207
+ },
208
+ {
209
+ "epoch": 4.5,
210
+ "eval_F1@10": 0.23374703381451323,
211
+ "eval_F1@5": 0.3258455720749842,
212
+ "eval_F1@M": 0.34501260196631756,
213
+ "eval_P@10": 0.2024,
214
+ "eval_P@5": 0.3964,
215
+ "eval_P@M": 0.4511738095238098,
216
+ "eval_R@10": 0.29999401709401724,
217
+ "eval_R@5": 0.2942325452325456,
218
+ "eval_R@M": 0.29999401709401724,
219
+ "eval_loss": 1.8626010417938232,
220
+ "eval_model_name": "t5-small-keyword-generation-inspec",
221
+ "eval_runtime": 281.3111,
222
+ "eval_samples_per_second": 1.777,
223
+ "eval_steps_per_second": 1.777,
224
+ "step": 4500
225
+ },
226
+ {
227
+ "epoch": 5.0,
228
+ "learning_rate": 9.185510204081633e-05,
229
+ "loss": 1.5068,
230
+ "step": 5000
231
+ },
232
+ {
233
+ "epoch": 5.0,
234
+ "eval_F1@10": 0.24335615627100407,
235
+ "eval_F1@5": 0.3396655599302659,
236
+ "eval_F1@M": 0.3602646628912888,
237
+ "eval_P@10": 0.21099999999999977,
238
+ "eval_P@5": 0.4139999999999996,
239
+ "eval_P@M": 0.4727857142857141,
240
+ "eval_R@10": 0.3119000777000777,
241
+ "eval_R@5": 0.30657915417915416,
242
+ "eval_R@M": 0.3119000777000777,
243
+ "eval_loss": 1.7979589700698853,
244
+ "eval_model_name": "t5-small-keyword-generation-inspec",
245
+ "eval_runtime": 280.6796,
246
+ "eval_samples_per_second": 1.781,
247
+ "eval_steps_per_second": 1.781,
248
+ "step": 5000
249
+ },
250
+ {
251
+ "epoch": 5.5,
252
+ "learning_rate": 9.083469387755103e-05,
253
+ "loss": 1.3146,
254
+ "step": 5500
255
+ },
256
+ {
257
+ "epoch": 5.5,
258
+ "eval_F1@10": 0.24001641670200757,
259
+ "eval_F1@5": 0.334409038673745,
260
+ "eval_F1@M": 0.3550115948035458,
261
+ "eval_P@10": 0.20839999999999997,
262
+ "eval_P@5": 0.40799999999999986,
263
+ "eval_P@M": 0.46737857142857153,
264
+ "eval_R@10": 0.30688550338550347,
265
+ "eval_R@5": 0.30161761016761024,
266
+ "eval_R@M": 0.30688550338550347,
267
+ "eval_loss": 1.8715476989746094,
268
+ "eval_model_name": "t5-small-keyword-generation-inspec",
269
+ "eval_runtime": 280.9732,
270
+ "eval_samples_per_second": 1.78,
271
+ "eval_steps_per_second": 1.78,
272
+ "step": 5500
273
+ },
274
+ {
275
+ "epoch": 6.0,
276
+ "learning_rate": 8.981428571428572e-05,
277
+ "loss": 1.2704,
278
+ "step": 6000
279
+ },
280
+ {
281
+ "epoch": 6.0,
282
+ "eval_F1@10": 0.24096180425219377,
283
+ "eval_F1@5": 0.334974600889307,
284
+ "eval_F1@M": 0.35426202280691466,
285
+ "eval_P@10": 0.20859999999999995,
286
+ "eval_P@5": 0.4076,
287
+ "eval_P@M": 0.4599071428571428,
288
+ "eval_R@10": 0.30986234321234324,
289
+ "eval_R@5": 0.30275404595404604,
290
+ "eval_R@M": 0.30986234321234324,
291
+ "eval_loss": 1.8688002824783325,
292
+ "eval_model_name": "t5-small-keyword-generation-inspec",
293
+ "eval_runtime": 279.0562,
294
+ "eval_samples_per_second": 1.792,
295
+ "eval_steps_per_second": 1.792,
296
+ "step": 6000
297
+ },
298
+ {
299
+ "epoch": 6.5,
300
+ "learning_rate": 8.879591836734694e-05,
301
+ "loss": 1.1731,
302
+ "step": 6500
303
+ },
304
+ {
305
+ "epoch": 6.5,
306
+ "eval_F1@10": 0.2441899110215107,
307
+ "eval_F1@5": 0.3420338693985756,
308
+ "eval_F1@M": 0.36420169024302196,
309
+ "eval_P@10": 0.2115999999999999,
310
+ "eval_P@5": 0.41599999999999987,
311
+ "eval_P@M": 0.4780619047619047,
312
+ "eval_R@10": 0.31397009102009105,
313
+ "eval_R@5": 0.3098475080475081,
314
+ "eval_R@M": 0.31397009102009105,
315
+ "eval_loss": 1.8536425828933716,
316
+ "eval_model_name": "t5-small-keyword-generation-inspec",
317
+ "eval_runtime": 291.8339,
318
+ "eval_samples_per_second": 1.713,
319
+ "eval_steps_per_second": 1.713,
320
+ "step": 6500
321
+ },
322
+ {
323
+ "epoch": 7.0,
324
+ "learning_rate": 8.777551020408164e-05,
325
+ "loss": 1.1174,
326
+ "step": 7000
327
+ },
328
+ {
329
+ "epoch": 7.0,
330
+ "eval_F1@10": 0.24262947404967916,
331
+ "eval_F1@5": 0.33894085685556324,
332
+ "eval_F1@M": 0.35926122996370713,
333
+ "eval_P@10": 0.21039999999999992,
334
+ "eval_P@5": 0.4128,
335
+ "eval_P@M": 0.46843333333333326,
336
+ "eval_R@10": 0.31144045954045946,
337
+ "eval_R@5": 0.30653605838605835,
338
+ "eval_R@M": 0.31144045954045946,
339
+ "eval_loss": 1.8547453880310059,
340
+ "eval_model_name": "t5-small-keyword-generation-inspec",
341
+ "eval_runtime": 293.7424,
342
+ "eval_samples_per_second": 1.702,
343
+ "eval_steps_per_second": 1.702,
344
+ "step": 7000
345
+ },
346
+ {
347
+ "epoch": 7.5,
348
+ "learning_rate": 8.675510204081633e-05,
349
+ "loss": 1.0319,
350
+ "step": 7500
351
+ },
352
+ {
353
+ "epoch": 7.5,
354
+ "eval_F1@10": 0.24435341427092733,
355
+ "eval_F1@5": 0.34094378497319694,
356
+ "eval_F1@M": 0.36301657635450263,
357
+ "eval_P@10": 0.21159999999999995,
358
+ "eval_P@5": 0.4139999999999999,
359
+ "eval_P@M": 0.4739333333333334,
360
+ "eval_R@10": 0.31494397269397284,
361
+ "eval_R@5": 0.30938704628704644,
362
+ "eval_R@M": 0.31494397269397284,
363
+ "eval_loss": 1.8866732120513916,
364
+ "eval_model_name": "t5-small-keyword-generation-inspec",
365
+ "eval_runtime": 298.9317,
366
+ "eval_samples_per_second": 1.673,
367
+ "eval_steps_per_second": 1.673,
368
+ "step": 7500
369
+ },
370
+ {
371
+ "epoch": 8.0,
372
+ "learning_rate": 8.573469387755101e-05,
373
+ "loss": 1.0172,
374
+ "step": 8000
375
+ },
376
+ {
377
+ "epoch": 8.0,
378
+ "eval_F1@10": 0.2461478679992482,
379
+ "eval_F1@5": 0.34169924781101274,
380
+ "eval_F1@M": 0.36424212479986534,
381
+ "eval_P@10": 0.21280000000000002,
382
+ "eval_P@5": 0.41439999999999994,
383
+ "eval_P@M": 0.47244523809523775,
384
+ "eval_R@10": 0.3185173215673219,
385
+ "eval_R@5": 0.3108509435009436,
386
+ "eval_R@M": 0.3185173215673219,
387
+ "eval_loss": 1.9575016498565674,
388
+ "eval_model_name": "t5-small-keyword-generation-inspec",
389
+ "eval_runtime": 264.7319,
390
+ "eval_samples_per_second": 1.889,
391
+ "eval_steps_per_second": 1.889,
392
+ "step": 8000
393
+ },
394
+ {
395
+ "epoch": 8.5,
396
+ "learning_rate": 8.471632653061225e-05,
397
+ "loss": 0.9324,
398
+ "step": 8500
399
+ },
400
+ {
401
+ "epoch": 8.5,
402
+ "eval_F1@10": 0.23868666530445426,
403
+ "eval_F1@5": 0.3324125028566208,
404
+ "eval_F1@M": 0.3539066449202676,
405
+ "eval_P@10": 0.20719999999999994,
406
+ "eval_P@5": 0.4051999999999999,
407
+ "eval_P@M": 0.46190476190476165,
408
+ "eval_R@10": 0.3066111166611168,
409
+ "eval_R@5": 0.3009154456654457,
410
+ "eval_R@M": 0.3066111166611168,
411
+ "eval_loss": 1.9488437175750732,
412
+ "eval_model_name": "t5-small-keyword-generation-inspec",
413
+ "eval_runtime": 286.0434,
414
+ "eval_samples_per_second": 1.748,
415
+ "eval_steps_per_second": 1.748,
416
+ "step": 8500
417
+ },
418
+ {
419
+ "epoch": 9.0,
420
+ "learning_rate": 8.369591836734695e-05,
421
+ "loss": 0.9587,
422
+ "step": 9000
423
+ },
424
+ {
425
+ "epoch": 9.0,
426
+ "eval_F1@10": 0.23996280126000177,
427
+ "eval_F1@5": 0.33370613928555143,
428
+ "eval_F1@M": 0.3559531203402726,
429
+ "eval_P@10": 0.20799999999999982,
430
+ "eval_P@5": 0.4059999999999997,
431
+ "eval_P@M": 0.4660785714285711,
432
+ "eval_R@10": 0.30781457986458,
433
+ "eval_R@5": 0.30168088578088587,
434
+ "eval_R@M": 0.30781457986458,
435
+ "eval_loss": 1.9363871812820435,
436
+ "eval_model_name": "t5-small-keyword-generation-inspec",
437
+ "eval_runtime": 282.9341,
438
+ "eval_samples_per_second": 1.767,
439
+ "eval_steps_per_second": 1.767,
440
+ "step": 9000
441
+ },
442
+ {
443
+ "epoch": 9.5,
444
+ "learning_rate": 8.267551020408164e-05,
445
+ "loss": 0.8864,
446
+ "step": 9500
447
+ },
448
+ {
449
+ "epoch": 9.5,
450
+ "eval_F1@10": 0.24553398899899156,
451
+ "eval_F1@5": 0.33974727821198425,
452
+ "eval_F1@M": 0.36148470868656674,
453
+ "eval_P@10": 0.21259999999999998,
454
+ "eval_P@5": 0.4128,
455
+ "eval_P@M": 0.46541666666666653,
456
+ "eval_R@10": 0.3148071484071486,
457
+ "eval_R@5": 0.3069217227217229,
458
+ "eval_R@M": 0.3148071484071486,
459
+ "eval_loss": 2.0047595500946045,
460
+ "eval_model_name": "t5-small-keyword-generation-inspec",
461
+ "eval_runtime": 283.5357,
462
+ "eval_samples_per_second": 1.763,
463
+ "eval_steps_per_second": 1.763,
464
+ "step": 9500
465
+ },
466
+ {
467
+ "epoch": 10.0,
468
+ "learning_rate": 8.165510204081634e-05,
469
+ "loss": 0.8219,
470
+ "step": 10000
471
+ },
472
+ {
473
+ "epoch": 10.0,
474
+ "eval_F1@10": 0.24542163861026467,
475
+ "eval_F1@5": 0.3405659428806491,
476
+ "eval_F1@M": 0.36243521048954536,
477
+ "eval_P@10": 0.2128,
478
+ "eval_P@5": 0.4147999999999997,
479
+ "eval_P@M": 0.471490476190476,
480
+ "eval_R@10": 0.31549298479298493,
481
+ "eval_R@5": 0.30797812187812196,
482
+ "eval_R@M": 0.31549298479298493,
483
+ "eval_loss": 2.050034999847412,
484
+ "eval_model_name": "t5-small-keyword-generation-inspec",
485
+ "eval_runtime": 284.698,
486
+ "eval_samples_per_second": 1.756,
487
+ "eval_steps_per_second": 1.756,
488
+ "step": 10000
489
+ },
490
+ {
491
+ "epoch": 10.5,
492
+ "learning_rate": 8.063469387755103e-05,
493
+ "loss": 0.7906,
494
+ "step": 10500
495
+ },
496
+ {
497
+ "epoch": 10.5,
498
+ "eval_F1@10": 0.24112739408629902,
499
+ "eval_F1@5": 0.3361905365875957,
500
+ "eval_F1@M": 0.35434909789213215,
501
+ "eval_P@10": 0.20859999999999984,
502
+ "eval_P@5": 0.4083999999999998,
503
+ "eval_P@M": 0.4588999999999998,
504
+ "eval_R@10": 0.3117027250527253,
505
+ "eval_R@5": 0.3052405316905319,
506
+ "eval_R@M": 0.3117027250527253,
507
+ "eval_loss": 2.076453685760498,
508
+ "eval_model_name": "t5-small-keyword-generation-inspec",
509
+ "eval_runtime": 283.1972,
510
+ "eval_samples_per_second": 1.766,
511
+ "eval_steps_per_second": 1.766,
512
+ "step": 10500
513
+ },
514
+ {
515
+ "epoch": 11.0,
516
+ "learning_rate": 7.961632653061225e-05,
517
+ "loss": 0.7828,
518
+ "step": 11000
519
+ },
520
+ {
521
+ "epoch": 11.0,
522
+ "eval_F1@10": 0.2434173157573352,
523
+ "eval_F1@5": 0.3369145302410012,
524
+ "eval_F1@M": 0.35818623186239346,
525
+ "eval_P@10": 0.21079999999999996,
526
+ "eval_P@5": 0.40959999999999974,
527
+ "eval_P@M": 0.4597690476190474,
528
+ "eval_R@10": 0.3140421911421914,
529
+ "eval_R@5": 0.3053971694971696,
530
+ "eval_R@M": 0.3140421911421914,
531
+ "eval_loss": 2.0442428588867188,
532
+ "eval_model_name": "t5-small-keyword-generation-inspec",
533
+ "eval_runtime": 283.6328,
534
+ "eval_samples_per_second": 1.763,
535
+ "eval_steps_per_second": 1.763,
536
+ "step": 11000
537
+ },
538
+ {
539
+ "epoch": 11.5,
540
+ "learning_rate": 7.859795918367348e-05,
541
+ "loss": 0.7724,
542
+ "step": 11500
543
+ },
544
+ {
545
+ "epoch": 11.5,
546
+ "eval_F1@10": 0.24332780944321591,
547
+ "eval_F1@5": 0.3358317365640898,
548
+ "eval_F1@M": 0.3594637432399048,
549
+ "eval_P@10": 0.21079999999999996,
550
+ "eval_P@5": 0.40799999999999986,
551
+ "eval_P@M": 0.4666904761904761,
552
+ "eval_R@10": 0.31255604395604414,
553
+ "eval_R@5": 0.3039292041292042,
554
+ "eval_R@M": 0.31255604395604414,
555
+ "eval_loss": 2.1262500286102295,
556
+ "eval_model_name": "t5-small-keyword-generation-inspec",
557
+ "eval_runtime": 283.3582,
558
+ "eval_samples_per_second": 1.765,
559
+ "eval_steps_per_second": 1.765,
560
+ "step": 11500
561
+ },
562
+ {
563
+ "epoch": 12.0,
564
+ "learning_rate": 7.757755102040817e-05,
565
+ "loss": 0.6985,
566
+ "step": 12000
567
+ },
568
+ {
569
+ "epoch": 12.0,
570
+ "eval_F1@10": 0.24282010444872945,
571
+ "eval_F1@5": 0.33782710328886834,
572
+ "eval_F1@M": 0.35967159486540345,
573
+ "eval_P@10": 0.2104,
574
+ "eval_P@5": 0.4112,
575
+ "eval_P@M": 0.4646380952380953,
576
+ "eval_R@10": 0.31213939948939956,
577
+ "eval_R@5": 0.30527879342879344,
578
+ "eval_R@M": 0.31213939948939956,
579
+ "eval_loss": 2.0867960453033447,
580
+ "eval_model_name": "t5-small-keyword-generation-inspec",
581
+ "eval_runtime": 281.5635,
582
+ "eval_samples_per_second": 1.776,
583
+ "eval_steps_per_second": 1.776,
584
+ "step": 12000
585
+ },
586
+ {
587
+ "epoch": 12.5,
588
+ "learning_rate": 7.655714285714286e-05,
589
+ "loss": 0.6647,
590
+ "step": 12500
591
+ },
592
+ {
593
+ "epoch": 12.5,
594
+ "eval_F1@10": 0.24765508121946467,
595
+ "eval_F1@5": 0.3433662775133368,
596
+ "eval_F1@M": 0.3637068867664847,
597
+ "eval_P@10": 0.21459999999999982,
598
+ "eval_P@5": 0.4179999999999996,
599
+ "eval_P@M": 0.4680285714285711,
600
+ "eval_R@10": 0.3187178987678989,
601
+ "eval_R@5": 0.3105277833277834,
602
+ "eval_R@M": 0.3187178987678989,
603
+ "eval_loss": 2.156297445297241,
604
+ "eval_model_name": "t5-small-keyword-generation-inspec",
605
+ "eval_runtime": 281.7625,
606
+ "eval_samples_per_second": 1.775,
607
+ "eval_steps_per_second": 1.775,
608
+ "step": 12500
609
+ },
610
+ {
611
+ "epoch": 13.0,
612
+ "learning_rate": 7.553673469387756e-05,
613
+ "loss": 0.6823,
614
+ "step": 13000
615
+ },
616
+ {
617
+ "epoch": 13.0,
618
+ "eval_F1@10": 0.25070757222690915,
619
+ "eval_F1@5": 0.34642384380325597,
620
+ "eval_F1@M": 0.36789754023066784,
621
+ "eval_P@10": 0.21759999999999993,
622
+ "eval_P@5": 0.4223999999999997,
623
+ "eval_P@M": 0.47199523809523813,
624
+ "eval_R@10": 0.3204831224331225,
625
+ "eval_R@5": 0.3122700632700633,
626
+ "eval_R@M": 0.3204831224331225,
627
+ "eval_loss": 2.154297113418579,
628
+ "eval_model_name": "t5-small-keyword-generation-inspec",
629
+ "eval_runtime": 283.3444,
630
+ "eval_samples_per_second": 1.765,
631
+ "eval_steps_per_second": 1.765,
632
+ "step": 13000
633
+ },
634
+ {
635
+ "epoch": 13.5,
636
+ "learning_rate": 7.451632653061224e-05,
637
+ "loss": 0.6039,
638
+ "step": 13500
639
+ },
640
+ {
641
+ "epoch": 13.5,
642
+ "eval_F1@10": 0.24724974591747426,
643
+ "eval_F1@5": 0.34381850469791675,
644
+ "eval_F1@M": 0.36370015266810973,
645
+ "eval_P@10": 0.21420000000000003,
646
+ "eval_P@5": 0.4179999999999999,
647
+ "eval_P@M": 0.4715571428571432,
648
+ "eval_R@10": 0.31813237873237904,
649
+ "eval_R@5": 0.31155258075258097,
650
+ "eval_R@M": 0.31813237873237904,
651
+ "eval_loss": 2.273547410964966,
652
+ "eval_model_name": "t5-small-keyword-generation-inspec",
653
+ "eval_runtime": 282.8458,
654
+ "eval_samples_per_second": 1.768,
655
+ "eval_steps_per_second": 1.768,
656
+ "step": 13500
657
+ },
658
+ {
659
+ "epoch": 14.0,
660
+ "learning_rate": 7.349591836734695e-05,
661
+ "loss": 0.6657,
662
+ "step": 14000
663
+ },
664
+ {
665
+ "epoch": 14.0,
666
+ "eval_F1@10": 0.24294545526488653,
667
+ "eval_F1@5": 0.33743259452965385,
668
+ "eval_F1@M": 0.35771661821151046,
669
+ "eval_P@10": 0.21059999999999995,
670
+ "eval_P@5": 0.4103999999999999,
671
+ "eval_P@M": 0.4632714285714285,
672
+ "eval_R@10": 0.3123259573759575,
673
+ "eval_R@5": 0.3057819458319459,
674
+ "eval_R@M": 0.3123259573759575,
675
+ "eval_loss": 2.2293343544006348,
676
+ "eval_model_name": "t5-small-keyword-generation-inspec",
677
+ "eval_runtime": 284.7498,
678
+ "eval_samples_per_second": 1.756,
679
+ "eval_steps_per_second": 1.756,
680
+ "step": 14000
681
+ },
682
+ {
683
+ "epoch": 14.5,
684
+ "learning_rate": 7.247551020408163e-05,
685
+ "loss": 0.5512,
686
+ "step": 14500
687
+ },
688
+ {
689
+ "epoch": 14.5,
690
+ "eval_F1@10": 0.24609090645971843,
691
+ "eval_F1@5": 0.341195668057433,
692
+ "eval_F1@M": 0.3638307651069263,
693
+ "eval_P@10": 0.2134,
694
+ "eval_P@5": 0.4151999999999999,
695
+ "eval_P@M": 0.47303809523809526,
696
+ "eval_R@10": 0.31571549561549583,
697
+ "eval_R@5": 0.3083380064380067,
698
+ "eval_R@M": 0.31571549561549583,
699
+ "eval_loss": 2.2570900917053223,
700
+ "eval_model_name": "t5-small-keyword-generation-inspec",
701
+ "eval_runtime": 289.2337,
702
+ "eval_samples_per_second": 1.729,
703
+ "eval_steps_per_second": 1.729,
704
+ "step": 14500
705
+ },
706
+ {
707
+ "epoch": 15.0,
708
+ "learning_rate": 7.145510204081633e-05,
709
+ "loss": 0.6125,
710
+ "step": 15000
711
+ },
712
+ {
713
+ "epoch": 15.0,
714
+ "eval_F1@10": 0.24506923303510375,
715
+ "eval_F1@5": 0.34101718379659585,
716
+ "eval_F1@M": 0.36054259558810675,
717
+ "eval_P@10": 0.21240000000000006,
718
+ "eval_P@5": 0.41520000000000007,
719
+ "eval_P@M": 0.46649523809523824,
720
+ "eval_R@10": 0.3145362748362751,
721
+ "eval_R@5": 0.30819146964146976,
722
+ "eval_R@M": 0.3145362748362751,
723
+ "eval_loss": 2.300617218017578,
724
+ "eval_model_name": "t5-small-keyword-generation-inspec",
725
+ "eval_runtime": 285.2317,
726
+ "eval_samples_per_second": 1.753,
727
+ "eval_steps_per_second": 1.753,
728
+ "step": 15000
729
+ },
730
+ {
731
+ "epoch": 15.5,
732
+ "learning_rate": 7.043673469387755e-05,
733
+ "loss": 0.5785,
734
+ "step": 15500
735
+ },
736
+ {
737
+ "epoch": 15.5,
738
+ "eval_F1@10": 0.2442251260644657,
739
+ "eval_F1@5": 0.33893644394820904,
740
+ "eval_F1@M": 0.35870379697701743,
741
+ "eval_P@10": 0.21159999999999998,
742
+ "eval_P@5": 0.41240000000000004,
743
+ "eval_P@M": 0.46347380952380973,
744
+ "eval_R@10": 0.3134413031413033,
745
+ "eval_R@5": 0.3057845210345211,
746
+ "eval_R@M": 0.3134413031413033,
747
+ "eval_loss": 2.338202953338623,
748
+ "eval_model_name": "t5-small-keyword-generation-inspec",
749
+ "eval_runtime": 289.9816,
750
+ "eval_samples_per_second": 1.724,
751
+ "eval_steps_per_second": 1.724,
752
+ "step": 15500
753
+ },
754
+ {
755
+ "epoch": 16.0,
756
+ "learning_rate": 6.941632653061226e-05,
757
+ "loss": 0.5428,
758
+ "step": 16000
759
+ },
760
+ {
761
+ "epoch": 16.0,
762
+ "eval_F1@10": 0.2432725372472316,
763
+ "eval_F1@5": 0.3373570716864839,
764
+ "eval_F1@M": 0.3582228556960761,
765
+ "eval_P@10": 0.21059999999999998,
766
+ "eval_P@5": 0.4095999999999998,
767
+ "eval_P@M": 0.46375,
768
+ "eval_R@10": 0.3134618159618162,
769
+ "eval_R@5": 0.30584154179154194,
770
+ "eval_R@M": 0.3134618159618162,
771
+ "eval_loss": 2.3253493309020996,
772
+ "eval_model_name": "t5-small-keyword-generation-inspec",
773
+ "eval_runtime": 289.4094,
774
+ "eval_samples_per_second": 1.728,
775
+ "eval_steps_per_second": 1.728,
776
+ "step": 16000
777
+ },
778
+ {
779
+ "epoch": 16.5,
780
+ "learning_rate": 6.839591836734694e-05,
781
+ "loss": 0.5194,
782
+ "step": 16500
783
+ },
784
+ {
785
+ "epoch": 16.5,
786
+ "eval_F1@10": 0.2355793420179682,
787
+ "eval_F1@5": 0.3278029424823546,
788
+ "eval_F1@M": 0.3482424748282339,
789
+ "eval_P@10": 0.20440000000000003,
790
+ "eval_P@5": 0.3991999999999999,
791
+ "eval_P@M": 0.4537833333333335,
792
+ "eval_R@10": 0.30187379287379296,
793
+ "eval_R@5": 0.2962392329892331,
794
+ "eval_R@M": 0.30187379287379296,
795
+ "eval_loss": 2.3528101444244385,
796
+ "eval_model_name": "t5-small-keyword-generation-inspec",
797
+ "eval_runtime": 290.7706,
798
+ "eval_samples_per_second": 1.72,
799
+ "eval_steps_per_second": 1.72,
800
+ "step": 16500
801
+ },
802
+ {
803
+ "epoch": 17.0,
804
+ "learning_rate": 6.737551020408163e-05,
805
+ "loss": 0.5194,
806
+ "step": 17000
807
+ },
808
+ {
809
+ "epoch": 17.0,
810
+ "eval_F1@10": 0.24069023813939067,
811
+ "eval_F1@5": 0.33479577971048596,
812
+ "eval_F1@M": 0.355847497067776,
813
+ "eval_P@10": 0.20800000000000002,
814
+ "eval_P@5": 0.40559999999999996,
815
+ "eval_P@M": 0.46179999999999993,
816
+ "eval_R@10": 0.311403085803086,
817
+ "eval_R@5": 0.3045922632922635,
818
+ "eval_R@M": 0.311403085803086,
819
+ "eval_loss": 2.3786723613739014,
820
+ "eval_model_name": "t5-small-keyword-generation-inspec",
821
+ "eval_runtime": 272.2042,
822
+ "eval_samples_per_second": 1.837,
823
+ "eval_steps_per_second": 1.837,
824
+ "step": 17000
825
+ },
826
+ {
827
+ "epoch": 17.5,
828
+ "learning_rate": 6.635510204081633e-05,
829
+ "loss": 0.4734,
830
+ "step": 17500
831
+ },
832
+ {
833
+ "epoch": 17.5,
834
+ "eval_F1@10": 0.2424905535962076,
835
+ "eval_F1@5": 0.33929873949579886,
836
+ "eval_F1@M": 0.36014400369152755,
837
+ "eval_P@10": 0.20979999999999996,
838
+ "eval_P@5": 0.4119999999999998,
839
+ "eval_P@M": 0.46821190476190494,
840
+ "eval_R@10": 0.3125397380397382,
841
+ "eval_R@5": 0.3077346153846155,
842
+ "eval_R@M": 0.3125397380397382,
843
+ "eval_loss": 2.3801848888397217,
844
+ "eval_model_name": "t5-small-keyword-generation-inspec",
845
+ "eval_runtime": 269.1685,
846
+ "eval_samples_per_second": 1.858,
847
+ "eval_steps_per_second": 1.858,
848
+ "step": 17500
849
+ },
850
+ {
851
+ "epoch": 18.0,
852
+ "learning_rate": 6.533469387755102e-05,
853
+ "loss": 0.4685,
854
+ "step": 18000
855
+ },
856
+ {
857
+ "epoch": 18.0,
858
+ "eval_F1@10": 0.2428809489937239,
859
+ "eval_F1@5": 0.34102215725451057,
860
+ "eval_F1@M": 0.36136990799018703,
861
+ "eval_P@10": 0.20919999999999994,
862
+ "eval_P@5": 0.41119999999999984,
863
+ "eval_P@M": 0.4681333333333333,
864
+ "eval_R@10": 0.31634897324897343,
865
+ "eval_R@5": 0.31194796314796325,
866
+ "eval_R@M": 0.31634897324897343,
867
+ "eval_loss": 2.4091179370880127,
868
+ "eval_model_name": "t5-small-keyword-generation-inspec",
869
+ "eval_runtime": 264.9477,
870
+ "eval_samples_per_second": 1.887,
871
+ "eval_steps_per_second": 1.887,
872
+ "step": 18000
873
+ },
874
+ {
875
+ "epoch": 18.5,
876
+ "learning_rate": 6.431428571428572e-05,
877
+ "loss": 0.4574,
878
+ "step": 18500
879
+ },
880
+ {
881
+ "epoch": 18.5,
882
+ "eval_F1@10": 0.24470615952450722,
883
+ "eval_F1@5": 0.34083279204455713,
884
+ "eval_F1@M": 0.3591474804177593,
885
+ "eval_P@10": 0.21199999999999997,
886
+ "eval_P@5": 0.41440000000000005,
887
+ "eval_P@M": 0.46359047619047605,
888
+ "eval_R@10": 0.31490344655344676,
889
+ "eval_R@5": 0.30879514929514945,
890
+ "eval_R@M": 0.31490344655344676,
891
+ "eval_loss": 2.414991617202759,
892
+ "eval_model_name": "t5-small-keyword-generation-inspec",
893
+ "eval_runtime": 289.5904,
894
+ "eval_samples_per_second": 1.727,
895
+ "eval_steps_per_second": 1.727,
896
+ "step": 18500
897
+ },
898
+ {
899
+ "epoch": 19.0,
900
+ "learning_rate": 6.329387755102041e-05,
901
+ "loss": 0.4411,
902
+ "step": 19000
903
+ },
904
+ {
905
+ "epoch": 19.0,
906
+ "eval_F1@10": 0.2413860020074456,
907
+ "eval_F1@5": 0.3365299011446076,
908
+ "eval_F1@M": 0.3552537179710558,
909
+ "eval_P@10": 0.20879999999999999,
910
+ "eval_P@5": 0.4083999999999998,
911
+ "eval_P@M": 0.4593714285714285,
912
+ "eval_R@10": 0.31052307137307145,
913
+ "eval_R@5": 0.3046235042735043,
914
+ "eval_R@M": 0.31052307137307145,
915
+ "eval_loss": 2.5369248390197754,
916
+ "eval_model_name": "t5-small-keyword-generation-inspec",
917
+ "eval_runtime": 281.7996,
918
+ "eval_samples_per_second": 1.774,
919
+ "eval_steps_per_second": 1.774,
920
+ "step": 19000
921
+ },
922
+ {
923
+ "epoch": 19.5,
924
+ "learning_rate": 6.22734693877551e-05,
925
+ "loss": 0.4032,
926
+ "step": 19500
927
+ },
928
+ {
929
+ "epoch": 19.5,
930
+ "eval_F1@10": 0.2432083462438766,
931
+ "eval_F1@5": 0.3403430030753566,
932
+ "eval_F1@M": 0.35755804044596656,
933
+ "eval_P@10": 0.2102,
934
+ "eval_P@5": 0.4132000000000002,
935
+ "eval_P@M": 0.4617999999999998,
936
+ "eval_R@10": 0.3141284104784106,
937
+ "eval_R@5": 0.30864017094017104,
938
+ "eval_R@M": 0.3141284104784106,
939
+ "eval_loss": 2.560804605484009,
940
+ "eval_model_name": "t5-small-keyword-generation-inspec",
941
+ "eval_runtime": 296.5123,
942
+ "eval_samples_per_second": 1.686,
943
+ "eval_steps_per_second": 1.686,
944
+ "step": 19500
945
+ },
946
+ {
947
+ "epoch": 20.0,
948
+ "learning_rate": 6.125510204081633e-05,
949
+ "loss": 0.4402,
950
+ "step": 20000
951
+ },
952
+ {
953
+ "epoch": 20.0,
954
+ "eval_F1@10": 0.24569117638243435,
955
+ "eval_F1@5": 0.3409062858709922,
956
+ "eval_F1@M": 0.3611119221851426,
957
+ "eval_P@10": 0.21239999999999998,
958
+ "eval_P@5": 0.41359999999999997,
959
+ "eval_P@M": 0.4668095238095235,
960
+ "eval_R@10": 0.3174460150960153,
961
+ "eval_R@5": 0.3095908924408926,
962
+ "eval_R@M": 0.3174460150960153,
963
+ "eval_loss": 2.5116209983825684,
964
+ "eval_model_name": "t5-small-keyword-generation-inspec",
965
+ "eval_runtime": 280.4633,
966
+ "eval_samples_per_second": 1.783,
967
+ "eval_steps_per_second": 1.783,
968
+ "step": 20000
969
+ }
970
+ ],
971
+ "max_steps": 50000,
972
+ "num_train_epochs": 50,
973
+ "total_flos": 2706836029440000.0,
974
+ "trial_name": null,
975
+ "trial_params": null
976
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ed43e6d8f4ec51ea2daf73c228ed358c1a0233bfe7c438145f1d47220364581b
3
+ size 3183