Hugging Face's logo

lamia24
/
archpipeline-evolution-ablation-directevolution

Safetensors
t5
archpipeline
software-architecture
adl
Model card Files
xet
 Community
lamia24 commited on
Commit
0814df3
·
verified ·
1 Parent(s): 18f0ee8

Upload folder using huggingface_hub

Browse files
Files changed (29) hide show
  1. checkpoint-1760/config.json +71 -0
  2. checkpoint-1760/generation_config.json +11 -0
  3. checkpoint-1760/model.safetensors +3 -0
  4. checkpoint-1760/optimizer.pt +3 -0
  5. checkpoint-1760/rng_state.pth +3 -0
  6. checkpoint-1760/scaler.pt +3 -0
  7. checkpoint-1760/scheduler.pt +3 -0
  8. checkpoint-1760/tokenizer.json +0 -0
  9. checkpoint-1760/tokenizer_config.json +122 -0
  10. checkpoint-1760/trainer_state.json +163 -0
  11. checkpoint-1760/training_args.bin +3 -0
  12. checkpoint-1980/config.json +71 -0
  13. checkpoint-1980/generation_config.json +11 -0
  14. checkpoint-1980/model.safetensors +3 -0
  15. checkpoint-1980/optimizer.pt +3 -0
  16. checkpoint-1980/rng_state.pth +3 -0
  17. checkpoint-1980/scaler.pt +3 -0
  18. checkpoint-1980/scheduler.pt +3 -0
  19. checkpoint-1980/tokenizer.json +0 -0
  20. checkpoint-1980/tokenizer_config.json +122 -0
  21. checkpoint-1980/trainer_state.json +178 -0
  22. checkpoint-1980/training_args.bin +3 -0
  23. config.json +71 -0
  24. generation_config.json +11 -0
  25. model.safetensors +3 -0
  26. resultats_Ablation_DirectEvolution_CodeT5base_FullFT.json +59 -0
  27. tokenizer.json +0 -0
  28. tokenizer_config.json +122 -0
  29. training_args.bin +3 -0
checkpoint-1760/config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5ForConditionalGeneration"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 3072,
8
+ "d_kv": 64,
9
+ "d_model": 768,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "dropout_rate": 0.1,
13
+ "dtype": "float32",
14
+ "eos_token_id": 2,
15
+ "feed_forward_proj": "relu",
16
+ "gradient_checkpointing": false,
17
+ "id2label": {
18
+ "0": "LABEL_0"
19
+ },
20
+ "initializer_factor": 1.0,
21
+ "is_decoder": false,
22
+ "is_encoder_decoder": true,
23
+ "is_gated_act": false,
24
+ "label2id": {
25
+ "LABEL_0": 0
26
+ },
27
+ "layer_norm_epsilon": 1e-06,
28
+ "model_type": "t5",
29
+ "n_positions": 512,
30
+ "num_decoder_layers": 12,
31
+ "num_heads": 12,
32
+ "num_layers": 12,
33
+ "output_past": true,
34
+ "pad_token_id": 0,
35
+ "relative_attention_max_distance": 128,
36
+ "relative_attention_num_buckets": 32,
37
+ "scale_decoder_outputs": true,
38
+ "task_specific_params": {
39
+ "summarization": {
40
+ "early_stopping": true,
41
+ "length_penalty": 2.0,
42
+ "max_length": 200,
43
+ "min_length": 30,
44
+ "no_repeat_ngram_size": 3,
45
+ "num_beams": 4,
46
+ "prefix": "summarize: "
47
+ },
48
+ "translation_en_to_de": {
49
+ "early_stopping": true,
50
+ "max_length": 300,
51
+ "num_beams": 4,
52
+ "prefix": "translate English to German: "
53
+ },
54
+ "translation_en_to_fr": {
55
+ "early_stopping": true,
56
+ "max_length": 300,
57
+ "num_beams": 4,
58
+ "prefix": "translate English to French: "
59
+ },
60
+ "translation_en_to_ro": {
61
+ "early_stopping": true,
62
+ "max_length": 300,
63
+ "num_beams": 4,
64
+ "prefix": "translate English to Romanian: "
65
+ }
66
+ },
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_cache": false,
70
+ "vocab_size": 32100
71
+ }
checkpoint-1760/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "decoder_start_token_id": 0,
5
+ "eos_token_id": [
6
+ 2
7
+ ],
8
+ "pad_token_id": 0,
9
+ "transformers_version": "5.0.0",
10
+ "use_cache": true
11
+ }
checkpoint-1760/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0dad66675633e5e49561d7a147d23d8926c4ca3216729c0b9c8e79312d8d4bf
3
+ size 1187392592
checkpoint-1760/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c9ca6b0773cd373b910b2542c6587d10ae22a90d6e9bee656051b82e5595b81
3
+ size 2177719755
checkpoint-1760/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c9c189dd00a45fb39c00e824c463b8ed365fd8db830fb552ec7eeca4bba3fa0f
3
+ size 14645
checkpoint-1760/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f7f2f965d2db5576ee3b15eb73913964e83bfa88d286b89a8af0f82b0923633f
3
+ size 1383
checkpoint-1760/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ba5fb987230a7591cb20d3aaada39d75c835f6b13cdd580ee20f3c9d2bb40049
3
+ size 1465
checkpoint-1760/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1760/tokenizer_config.json ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<extra_id_99>",
10
+ "<extra_id_98>",
11
+ "<extra_id_97>",
12
+ "<extra_id_96>",
13
+ "<extra_id_95>",
14
+ "<extra_id_94>",
15
+ "<extra_id_93>",
16
+ "<extra_id_92>",
17
+ "<extra_id_91>",
18
+ "<extra_id_90>",
19
+ "<extra_id_89>",
20
+ "<extra_id_88>",
21
+ "<extra_id_87>",
22
+ "<extra_id_86>",
23
+ "<extra_id_85>",
24
+ "<extra_id_84>",
25
+ "<extra_id_83>",
26
+ "<extra_id_82>",
27
+ "<extra_id_81>",
28
+ "<extra_id_80>",
29
+ "<extra_id_79>",
30
+ "<extra_id_78>",
31
+ "<extra_id_77>",
32
+ "<extra_id_76>",
33
+ "<extra_id_75>",
34
+ "<extra_id_74>",
35
+ "<extra_id_73>",
36
+ "<extra_id_72>",
37
+ "<extra_id_71>",
38
+ "<extra_id_70>",
39
+ "<extra_id_69>",
40
+ "<extra_id_68>",
41
+ "<extra_id_67>",
42
+ "<extra_id_66>",
43
+ "<extra_id_65>",
44
+ "<extra_id_64>",
45
+ "<extra_id_63>",
46
+ "<extra_id_62>",
47
+ "<extra_id_61>",
48
+ "<extra_id_60>",
49
+ "<extra_id_59>",
50
+ "<extra_id_58>",
51
+ "<extra_id_57>",
52
+ "<extra_id_56>",
53
+ "<extra_id_55>",
54
+ "<extra_id_54>",
55
+ "<extra_id_53>",
56
+ "<extra_id_52>",
57
+ "<extra_id_51>",
58
+ "<extra_id_50>",
59
+ "<extra_id_49>",
60
+ "<extra_id_48>",
61
+ "<extra_id_47>",
62
+ "<extra_id_46>",
63
+ "<extra_id_45>",
64
+ "<extra_id_44>",
65
+ "<extra_id_43>",
66
+ "<extra_id_42>",
67
+ "<extra_id_41>",
68
+ "<extra_id_40>",
69
+ "<extra_id_39>",
70
+ "<extra_id_38>",
71
+ "<extra_id_37>",
72
+ "<extra_id_36>",
73
+ "<extra_id_35>",
74
+ "<extra_id_34>",
75
+ "<extra_id_33>",
76
+ "<extra_id_32>",
77
+ "<extra_id_31>",
78
+ "<extra_id_30>",
79
+ "<extra_id_29>",
80
+ "<extra_id_28>",
81
+ "<extra_id_27>",
82
+ "<extra_id_26>",
83
+ "<extra_id_25>",
84
+ "<extra_id_24>",
85
+ "<extra_id_23>",
86
+ "<extra_id_22>",
87
+ "<extra_id_21>",
88
+ "<extra_id_20>",
89
+ "<extra_id_19>",
90
+ "<extra_id_18>",
91
+ "<extra_id_17>",
92
+ "<extra_id_16>",
93
+ "<extra_id_15>",
94
+ "<extra_id_14>",
95
+ "<extra_id_13>",
96
+ "<extra_id_12>",
97
+ "<extra_id_11>",
98
+ "<extra_id_10>",
99
+ "<extra_id_9>",
100
+ "<extra_id_8>",
101
+ "<extra_id_7>",
102
+ "<extra_id_6>",
103
+ "<extra_id_5>",
104
+ "<extra_id_4>",
105
+ "<extra_id_3>",
106
+ "<extra_id_2>",
107
+ "<extra_id_1>",
108
+ "<extra_id_0>"
109
+ ],
110
+ "is_local": true,
111
+ "mask_token": "<mask>",
112
+ "max_length": 128,
113
+ "model_max_length": 512,
114
+ "pad_token": "<pad>",
115
+ "sep_token": "</s>",
116
+ "stride": 0,
117
+ "tokenizer_class": "TokenizersBackend",
118
+ "trim_offsets": true,
119
+ "truncation_side": "right",
120
+ "truncation_strategy": "longest_first",
121
+ "unk_token": "<unk>"
122
+ }
checkpoint-1760/trainer_state.json ADDED
@@ -0,0 +1,163 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1760,
3
+ "best_metric": 0.004073738120496273,
4
+ "best_model_checkpoint": "/content/drive/MyDrive/evo_ablation_DirectEvolution_CodeT5base_FullFT/checkpoint-1760",
5
+ "epoch": 8.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1760,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 1.0355877876281738,
15
+ "learning_rate": 4.991027181875014e-05,
16
+ "loss": 1.8348119562322442,
17
+ "step": 220
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_loss": 0.03580193594098091,
22
+ "eval_runtime": 3.039,
23
+ "eval_samples_per_second": 72.392,
24
+ "eval_steps_per_second": 18.098,
25
+ "step": 220
26
+ },
27
+ {
28
+ "epoch": 2.0,
29
+ "grad_norm": 0.07191833108663559,
30
+ "learning_rate": 4.925493631367695e-05,
31
+ "loss": 0.08348724191839045,
32
+ "step": 440
33
+ },
34
+ {
35
+ "epoch": 2.0,
36
+ "eval_loss": 0.019032808020710945,
37
+ "eval_runtime": 3.0215,
38
+ "eval_samples_per_second": 72.811,
39
+ "eval_steps_per_second": 18.203,
40
+ "step": 440
41
+ },
42
+ {
43
+ "epoch": 3.0,
44
+ "grad_norm": 1.5394785404205322,
45
+ "learning_rate": 4.79727833666922e-05,
46
+ "loss": 0.04584090492942117,
47
+ "step": 660
48
+ },
49
+ {
50
+ "epoch": 3.0,
51
+ "eval_loss": 0.012234838679432869,
52
+ "eval_runtime": 3.0514,
53
+ "eval_samples_per_second": 72.098,
54
+ "eval_steps_per_second": 18.024,
55
+ "step": 660
56
+ },
57
+ {
58
+ "epoch": 4.0,
59
+ "grad_norm": 0.1377420276403427,
60
+ "learning_rate": 4.610819813755038e-05,
61
+ "loss": 0.025517858158458362,
62
+ "step": 880
63
+ },
64
+ {
65
+ "epoch": 4.0,
66
+ "eval_loss": 0.00837874785065651,
67
+ "eval_runtime": 3.0383,
68
+ "eval_samples_per_second": 72.409,
69
+ "eval_steps_per_second": 18.102,
70
+ "step": 880
71
+ },
72
+ {
73
+ "epoch": 5.0,
74
+ "grad_norm": 0.66103595495224,
75
+ "learning_rate": 4.3692257202498335e-05,
76
+ "loss": 0.017070700905539773,
77
+ "step": 1100
78
+ },
79
+ {
80
+ "epoch": 5.0,
81
+ "eval_loss": 0.006502046715468168,
82
+ "eval_runtime": 3.2267,
83
+ "eval_samples_per_second": 68.182,
84
+ "eval_steps_per_second": 17.045,
85
+ "step": 1100
86
+ },
87
+ {
88
+ "epoch": 6.0,
89
+ "grad_norm": 0.002123212441802025,
90
+ "learning_rate": 4.079444088124431e-05,
91
+ "loss": 0.012019785967740145,
92
+ "step": 1320
93
+ },
94
+ {
95
+ "epoch": 6.0,
96
+ "eval_loss": 0.00527948047965765,
97
+ "eval_runtime": 3.1399,
98
+ "eval_samples_per_second": 70.066,
99
+ "eval_steps_per_second": 17.516,
100
+ "step": 1320
101
+ },
102
+ {
103
+ "epoch": 7.0,
104
+ "grad_norm": 0.03506750613451004,
105
+ "learning_rate": 3.7489453172552035e-05,
106
+ "loss": 0.008358093283393167,
107
+ "step": 1540
108
+ },
109
+ {
110
+ "epoch": 7.0,
111
+ "eval_loss": 0.004877733998000622,
112
+ "eval_runtime": 3.0578,
113
+ "eval_samples_per_second": 71.948,
114
+ "eval_steps_per_second": 17.987,
115
+ "step": 1540
116
+ },
117
+ {
118
+ "epoch": 8.0,
119
+ "grad_norm": 0.0033804229460656643,
120
+ "learning_rate": 3.386249471384405e-05,
121
+ "loss": 0.007093059474771673,
122
+ "step": 1760
123
+ },
124
+ {
125
+ "epoch": 8.0,
126
+ "eval_loss": 0.004073738120496273,
127
+ "eval_runtime": 3.0231,
128
+ "eval_samples_per_second": 72.772,
129
+ "eval_steps_per_second": 18.193,
130
+ "step": 1760
131
+ }
132
+ ],
133
+ "logging_steps": 500,
134
+ "max_steps": 4400,
135
+ "num_input_tokens_seen": 0,
136
+ "num_train_epochs": 20,
137
+ "save_steps": 500,
138
+ "stateful_callbacks": {
139
+ "EarlyStoppingCallback": {
140
+ "args": {
141
+ "early_stopping_patience": 3,
142
+ "early_stopping_threshold": 0.001
143
+ },
144
+ "attributes": {
145
+ "early_stopping_patience_counter": 2
146
+ }
147
+ },
148
+ "TrainerControl": {
149
+ "args": {
150
+ "should_epoch_stop": false,
151
+ "should_evaluate": false,
152
+ "should_log": false,
153
+ "should_save": true,
154
+ "should_training_stop": false
155
+ },
156
+ "attributes": {}
157
+ }
158
+ },
159
+ "total_flos": 3188462920519680.0,
160
+ "train_batch_size": 4,
161
+ "trial_name": null,
162
+ "trial_params": null
163
+ }
checkpoint-1760/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b35a5e7fd07656f48c5e4b636b892270a761fa6de6c0e189afe38d1b4b0a8ba8
3
+ size 5393
checkpoint-1980/config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5ForConditionalGeneration"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 3072,
8
+ "d_kv": 64,
9
+ "d_model": 768,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "dropout_rate": 0.1,
13
+ "dtype": "float32",
14
+ "eos_token_id": 2,
15
+ "feed_forward_proj": "relu",
16
+ "gradient_checkpointing": false,
17
+ "id2label": {
18
+ "0": "LABEL_0"
19
+ },
20
+ "initializer_factor": 1.0,
21
+ "is_decoder": false,
22
+ "is_encoder_decoder": true,
23
+ "is_gated_act": false,
24
+ "label2id": {
25
+ "LABEL_0": 0
26
+ },
27
+ "layer_norm_epsilon": 1e-06,
28
+ "model_type": "t5",
29
+ "n_positions": 512,
30
+ "num_decoder_layers": 12,
31
+ "num_heads": 12,
32
+ "num_layers": 12,
33
+ "output_past": true,
34
+ "pad_token_id": 0,
35
+ "relative_attention_max_distance": 128,
36
+ "relative_attention_num_buckets": 32,
37
+ "scale_decoder_outputs": true,
38
+ "task_specific_params": {
39
+ "summarization": {
40
+ "early_stopping": true,
41
+ "length_penalty": 2.0,
42
+ "max_length": 200,
43
+ "min_length": 30,
44
+ "no_repeat_ngram_size": 3,
45
+ "num_beams": 4,
46
+ "prefix": "summarize: "
47
+ },
48
+ "translation_en_to_de": {
49
+ "early_stopping": true,
50
+ "max_length": 300,
51
+ "num_beams": 4,
52
+ "prefix": "translate English to German: "
53
+ },
54
+ "translation_en_to_fr": {
55
+ "early_stopping": true,
56
+ "max_length": 300,
57
+ "num_beams": 4,
58
+ "prefix": "translate English to French: "
59
+ },
60
+ "translation_en_to_ro": {
61
+ "early_stopping": true,
62
+ "max_length": 300,
63
+ "num_beams": 4,
64
+ "prefix": "translate English to Romanian: "
65
+ }
66
+ },
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_cache": false,
70
+ "vocab_size": 32100
71
+ }
checkpoint-1980/generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "decoder_start_token_id": 0,
5
+ "eos_token_id": [
6
+ 2
7
+ ],
8
+ "pad_token_id": 0,
9
+ "transformers_version": "5.0.0",
10
+ "use_cache": true
11
+ }
checkpoint-1980/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8f845a9c707839e30f31bb2fffb03524b176ef1818fdffd2731bd526397b7ffd
3
+ size 1187392592
checkpoint-1980/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e734fc16ee89f729e207bc0d6ab8c403957a39bac04449543e1d3132ae34df5c
3
+ size 2177719755
checkpoint-1980/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa6c72d0c27f6309079cf173fca7cfc436c7b6f90a2830f7f894017d54538c39
3
+ size 14645
checkpoint-1980/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f74185ee579ba16007a8fb03aa031516b0c0cc5188837cac6c604d2d0853c40f
3
+ size 1383
checkpoint-1980/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a7553c541c2a0e694f41560ab6ff6c4087bdf0ed00ac5d6b9cb70ccb131e1694
3
+ size 1465
checkpoint-1980/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
checkpoint-1980/tokenizer_config.json ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<extra_id_99>",
10
+ "<extra_id_98>",
11
+ "<extra_id_97>",
12
+ "<extra_id_96>",
13
+ "<extra_id_95>",
14
+ "<extra_id_94>",
15
+ "<extra_id_93>",
16
+ "<extra_id_92>",
17
+ "<extra_id_91>",
18
+ "<extra_id_90>",
19
+ "<extra_id_89>",
20
+ "<extra_id_88>",
21
+ "<extra_id_87>",
22
+ "<extra_id_86>",
23
+ "<extra_id_85>",
24
+ "<extra_id_84>",
25
+ "<extra_id_83>",
26
+ "<extra_id_82>",
27
+ "<extra_id_81>",
28
+ "<extra_id_80>",
29
+ "<extra_id_79>",
30
+ "<extra_id_78>",
31
+ "<extra_id_77>",
32
+ "<extra_id_76>",
33
+ "<extra_id_75>",
34
+ "<extra_id_74>",
35
+ "<extra_id_73>",
36
+ "<extra_id_72>",
37
+ "<extra_id_71>",
38
+ "<extra_id_70>",
39
+ "<extra_id_69>",
40
+ "<extra_id_68>",
41
+ "<extra_id_67>",
42
+ "<extra_id_66>",
43
+ "<extra_id_65>",
44
+ "<extra_id_64>",
45
+ "<extra_id_63>",
46
+ "<extra_id_62>",
47
+ "<extra_id_61>",
48
+ "<extra_id_60>",
49
+ "<extra_id_59>",
50
+ "<extra_id_58>",
51
+ "<extra_id_57>",
52
+ "<extra_id_56>",
53
+ "<extra_id_55>",
54
+ "<extra_id_54>",
55
+ "<extra_id_53>",
56
+ "<extra_id_52>",
57
+ "<extra_id_51>",
58
+ "<extra_id_50>",
59
+ "<extra_id_49>",
60
+ "<extra_id_48>",
61
+ "<extra_id_47>",
62
+ "<extra_id_46>",
63
+ "<extra_id_45>",
64
+ "<extra_id_44>",
65
+ "<extra_id_43>",
66
+ "<extra_id_42>",
67
+ "<extra_id_41>",
68
+ "<extra_id_40>",
69
+ "<extra_id_39>",
70
+ "<extra_id_38>",
71
+ "<extra_id_37>",
72
+ "<extra_id_36>",
73
+ "<extra_id_35>",
74
+ "<extra_id_34>",
75
+ "<extra_id_33>",
76
+ "<extra_id_32>",
77
+ "<extra_id_31>",
78
+ "<extra_id_30>",
79
+ "<extra_id_29>",
80
+ "<extra_id_28>",
81
+ "<extra_id_27>",
82
+ "<extra_id_26>",
83
+ "<extra_id_25>",
84
+ "<extra_id_24>",
85
+ "<extra_id_23>",
86
+ "<extra_id_22>",
87
+ "<extra_id_21>",
88
+ "<extra_id_20>",
89
+ "<extra_id_19>",
90
+ "<extra_id_18>",
91
+ "<extra_id_17>",
92
+ "<extra_id_16>",
93
+ "<extra_id_15>",
94
+ "<extra_id_14>",
95
+ "<extra_id_13>",
96
+ "<extra_id_12>",
97
+ "<extra_id_11>",
98
+ "<extra_id_10>",
99
+ "<extra_id_9>",
100
+ "<extra_id_8>",
101
+ "<extra_id_7>",
102
+ "<extra_id_6>",
103
+ "<extra_id_5>",
104
+ "<extra_id_4>",
105
+ "<extra_id_3>",
106
+ "<extra_id_2>",
107
+ "<extra_id_1>",
108
+ "<extra_id_0>"
109
+ ],
110
+ "is_local": true,
111
+ "mask_token": "<mask>",
112
+ "max_length": 128,
113
+ "model_max_length": 512,
114
+ "pad_token": "<pad>",
115
+ "sep_token": "</s>",
116
+ "stride": 0,
117
+ "tokenizer_class": "TokenizersBackend",
118
+ "trim_offsets": true,
119
+ "truncation_side": "right",
120
+ "truncation_strategy": "longest_first",
121
+ "unk_token": "<unk>"
122
+ }
checkpoint-1980/trainer_state.json ADDED
@@ -0,0 +1,178 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_global_step": 1760,
3
+ "best_metric": 0.004073738120496273,
4
+ "best_model_checkpoint": "/content/drive/MyDrive/evo_ablation_DirectEvolution_CodeT5base_FullFT/checkpoint-1760",
5
+ "epoch": 9.0,
6
+ "eval_steps": 500,
7
+ "global_step": 1980,
8
+ "is_hyper_param_search": false,
9
+ "is_local_process_zero": true,
10
+ "is_world_process_zero": true,
11
+ "log_history": [
12
+ {
13
+ "epoch": 1.0,
14
+ "grad_norm": 1.0355877876281738,
15
+ "learning_rate": 4.991027181875014e-05,
16
+ "loss": 1.8348119562322442,
17
+ "step": 220
18
+ },
19
+ {
20
+ "epoch": 1.0,
21
+ "eval_loss": 0.03580193594098091,
22
+ "eval_runtime": 3.039,
23
+ "eval_samples_per_second": 72.392,
24
+ "eval_steps_per_second": 18.098,
25
+ "step": 220
26
+ },
27
+ {
28
+ "epoch": 2.0,
29
+ "grad_norm": 0.07191833108663559,
30
+ "learning_rate": 4.925493631367695e-05,
31
+ "loss": 0.08348724191839045,
32
+ "step": 440
33
+ },
34
+ {
35
+ "epoch": 2.0,
36
+ "eval_loss": 0.019032808020710945,
37
+ "eval_runtime": 3.0215,
38
+ "eval_samples_per_second": 72.811,
39
+ "eval_steps_per_second": 18.203,
40
+ "step": 440
41
+ },
42
+ {
43
+ "epoch": 3.0,
44
+ "grad_norm": 1.5394785404205322,
45
+ "learning_rate": 4.79727833666922e-05,
46
+ "loss": 0.04584090492942117,
47
+ "step": 660
48
+ },
49
+ {
50
+ "epoch": 3.0,
51
+ "eval_loss": 0.012234838679432869,
52
+ "eval_runtime": 3.0514,
53
+ "eval_samples_per_second": 72.098,
54
+ "eval_steps_per_second": 18.024,
55
+ "step": 660
56
+ },
57
+ {
58
+ "epoch": 4.0,
59
+ "grad_norm": 0.1377420276403427,
60
+ "learning_rate": 4.610819813755038e-05,
61
+ "loss": 0.025517858158458362,
62
+ "step": 880
63
+ },
64
+ {
65
+ "epoch": 4.0,
66
+ "eval_loss": 0.00837874785065651,
67
+ "eval_runtime": 3.0383,
68
+ "eval_samples_per_second": 72.409,
69
+ "eval_steps_per_second": 18.102,
70
+ "step": 880
71
+ },
72
+ {
73
+ "epoch": 5.0,
74
+ "grad_norm": 0.66103595495224,
75
+ "learning_rate": 4.3692257202498335e-05,
76
+ "loss": 0.017070700905539773,
77
+ "step": 1100
78
+ },
79
+ {
80
+ "epoch": 5.0,
81
+ "eval_loss": 0.006502046715468168,
82
+ "eval_runtime": 3.2267,
83
+ "eval_samples_per_second": 68.182,
84
+ "eval_steps_per_second": 17.045,
85
+ "step": 1100
86
+ },
87
+ {
88
+ "epoch": 6.0,
89
+ "grad_norm": 0.002123212441802025,
90
+ "learning_rate": 4.079444088124431e-05,
91
+ "loss": 0.012019785967740145,
92
+ "step": 1320
93
+ },
94
+ {
95
+ "epoch": 6.0,
96
+ "eval_loss": 0.00527948047965765,
97
+ "eval_runtime": 3.1399,
98
+ "eval_samples_per_second": 70.066,
99
+ "eval_steps_per_second": 17.516,
100
+ "step": 1320
101
+ },
102
+ {
103
+ "epoch": 7.0,
104
+ "grad_norm": 0.03506750613451004,
105
+ "learning_rate": 3.7489453172552035e-05,
106
+ "loss": 0.008358093283393167,
107
+ "step": 1540
108
+ },
109
+ {
110
+ "epoch": 7.0,
111
+ "eval_loss": 0.004877733998000622,
112
+ "eval_runtime": 3.0578,
113
+ "eval_samples_per_second": 71.948,
114
+ "eval_steps_per_second": 17.987,
115
+ "step": 1540
116
+ },
117
+ {
118
+ "epoch": 8.0,
119
+ "grad_norm": 0.0033804229460656643,
120
+ "learning_rate": 3.386249471384405e-05,
121
+ "loss": 0.007093059474771673,
122
+ "step": 1760
123
+ },
124
+ {
125
+ "epoch": 8.0,
126
+ "eval_loss": 0.004073738120496273,
127
+ "eval_runtime": 3.0231,
128
+ "eval_samples_per_second": 72.772,
129
+ "eval_steps_per_second": 18.193,
130
+ "step": 1760
131
+ },
132
+ {
133
+ "epoch": 9.0,
134
+ "grad_norm": 0.0007005213410593569,
135
+ "learning_rate": 3.0007066358928138e-05,
136
+ "loss": 0.004284547404809431,
137
+ "step": 1980
138
+ },
139
+ {
140
+ "epoch": 9.0,
141
+ "eval_loss": 0.005608182400465012,
142
+ "eval_runtime": 3.0645,
143
+ "eval_samples_per_second": 71.789,
144
+ "eval_steps_per_second": 17.947,
145
+ "step": 1980
146
+ }
147
+ ],
148
+ "logging_steps": 500,
149
+ "max_steps": 4400,
150
+ "num_input_tokens_seen": 0,
151
+ "num_train_epochs": 20,
152
+ "save_steps": 500,
153
+ "stateful_callbacks": {
154
+ "EarlyStoppingCallback": {
155
+ "args": {
156
+ "early_stopping_patience": 3,
157
+ "early_stopping_threshold": 0.001
158
+ },
159
+ "attributes": {
160
+ "early_stopping_patience_counter": 3
161
+ }
162
+ },
163
+ "TrainerControl": {
164
+ "args": {
165
+ "should_epoch_stop": false,
166
+ "should_evaluate": false,
167
+ "should_log": false,
168
+ "should_save": true,
169
+ "should_training_stop": true
170
+ },
171
+ "attributes": {}
172
+ }
173
+ },
174
+ "total_flos": 3589231990763520.0,
175
+ "train_batch_size": 4,
176
+ "trial_name": null,
177
+ "trial_params": null
178
+ }
checkpoint-1980/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b35a5e7fd07656f48c5e4b636b892270a761fa6de6c0e189afe38d1b4b0a8ba8
3
+ size 5393
config.json ADDED
@@ -0,0 +1,71 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "T5ForConditionalGeneration"
4
+ ],
5
+ "bos_token_id": 1,
6
+ "classifier_dropout": 0.0,
7
+ "d_ff": 3072,
8
+ "d_kv": 64,
9
+ "d_model": 768,
10
+ "decoder_start_token_id": 0,
11
+ "dense_act_fn": "relu",
12
+ "dropout_rate": 0.1,
13
+ "dtype": "float32",
14
+ "eos_token_id": 2,
15
+ "feed_forward_proj": "relu",
16
+ "gradient_checkpointing": false,
17
+ "id2label": {
18
+ "0": "LABEL_0"
19
+ },
20
+ "initializer_factor": 1.0,
21
+ "is_decoder": false,
22
+ "is_encoder_decoder": true,
23
+ "is_gated_act": false,
24
+ "label2id": {
25
+ "LABEL_0": 0
26
+ },
27
+ "layer_norm_epsilon": 1e-06,
28
+ "model_type": "t5",
29
+ "n_positions": 512,
30
+ "num_decoder_layers": 12,
31
+ "num_heads": 12,
32
+ "num_layers": 12,
33
+ "output_past": true,
34
+ "pad_token_id": 0,
35
+ "relative_attention_max_distance": 128,
36
+ "relative_attention_num_buckets": 32,
37
+ "scale_decoder_outputs": true,
38
+ "task_specific_params": {
39
+ "summarization": {
40
+ "early_stopping": true,
41
+ "length_penalty": 2.0,
42
+ "max_length": 200,
43
+ "min_length": 30,
44
+ "no_repeat_ngram_size": 3,
45
+ "num_beams": 4,
46
+ "prefix": "summarize: "
47
+ },
48
+ "translation_en_to_de": {
49
+ "early_stopping": true,
50
+ "max_length": 300,
51
+ "num_beams": 4,
52
+ "prefix": "translate English to German: "
53
+ },
54
+ "translation_en_to_fr": {
55
+ "early_stopping": true,
56
+ "max_length": 300,
57
+ "num_beams": 4,
58
+ "prefix": "translate English to French: "
59
+ },
60
+ "translation_en_to_ro": {
61
+ "early_stopping": true,
62
+ "max_length": 300,
63
+ "num_beams": 4,
64
+ "prefix": "translate English to Romanian: "
65
+ }
66
+ },
67
+ "tie_word_embeddings": true,
68
+ "transformers_version": "5.0.0",
69
+ "use_cache": false,
70
+ "vocab_size": 32100
71
+ }
generation_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 1,
4
+ "decoder_start_token_id": 0,
5
+ "eos_token_id": [
6
+ 2
7
+ ],
8
+ "pad_token_id": 0,
9
+ "transformers_version": "5.0.0",
10
+ "use_cache": true
11
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0dad66675633e5e49561d7a147d23d8926c4ca3216729c0b9c8e79312d8d4bf
3
+ size 1187392592
resultats_Ablation_DirectEvolution_CodeT5base_FullFT.json ADDED
@@ -0,0 +1,59 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "modele": "Ablation_DirectEvolution_CodeT5base_FullFT",
3
+ "nb_test": 120,
4
+ "EM": 0.325,
5
+ "BLEU": 0.582636058188466,
6
+ "Smooth_BLEU": 0.582636058188466,
7
+ "ROUGE-1": 0.925754730250051,
8
+ "ROUGE-2": 0.8869814537034098,
9
+ "ROUGE-L": 0.9227087533582736,
10
+ "METEOR": 0.8832261012398559,
11
+ "BERT-P": 0.9993981719017029,
12
+ "BERT-R": 0.9993981719017029,
13
+ "BERT-F1": 0.9993981719017029,
14
+ "validite_syntaxique": {
15
+ "validite_globale": 0.9416666666666667,
16
+ "validite_ACME": 1.0,
17
+ "validite_AADL": 0.8653846153846154,
18
+ "nb_valides": 113,
19
+ "nb_total": 120
20
+ },
21
+ "precision_operationnelle": {
22
+ "ADD_CONNECTOR": {
23
+ "correct": 20,
24
+ "total": 20,
25
+ "precision": 1.0
26
+ },
27
+ "DELETE_CONNECTOR": {
28
+ "correct": 20,
29
+ "total": 20,
30
+ "precision": 1.0
31
+ },
32
+ "MODIFY_PROPERTY": {
33
+ "correct": 20,
34
+ "total": 20,
35
+ "precision": 1.0
36
+ },
37
+ "ADD_COMPONENT": {
38
+ "correct": 9,
39
+ "total": 20,
40
+ "precision": 0.45
41
+ },
42
+ "ADD_PORT": {
43
+ "correct": 19,
44
+ "total": 20,
45
+ "precision": 0.95
46
+ },
47
+ "DELETE_COMPONENT": {
48
+ "correct": 20,
49
+ "total": 20,
50
+ "precision": 1.0
51
+ },
52
+ "GLOBAL": {
53
+ "correct": 108,
54
+ "total": 120,
55
+ "precision": 0.9
56
+ }
57
+ },
58
+ "score_composite": 0.9364
59
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "backend": "tokenizers",
4
+ "bos_token": "<s>",
5
+ "cls_token": "<s>",
6
+ "eos_token": "</s>",
7
+ "errors": "replace",
8
+ "extra_special_tokens": [
9
+ "<extra_id_99>",
10
+ "<extra_id_98>",
11
+ "<extra_id_97>",
12
+ "<extra_id_96>",
13
+ "<extra_id_95>",
14
+ "<extra_id_94>",
15
+ "<extra_id_93>",
16
+ "<extra_id_92>",
17
+ "<extra_id_91>",
18
+ "<extra_id_90>",
19
+ "<extra_id_89>",
20
+ "<extra_id_88>",
21
+ "<extra_id_87>",
22
+ "<extra_id_86>",
23
+ "<extra_id_85>",
24
+ "<extra_id_84>",
25
+ "<extra_id_83>",
26
+ "<extra_id_82>",
27
+ "<extra_id_81>",
28
+ "<extra_id_80>",
29
+ "<extra_id_79>",
30
+ "<extra_id_78>",
31
+ "<extra_id_77>",
32
+ "<extra_id_76>",
33
+ "<extra_id_75>",
34
+ "<extra_id_74>",
35
+ "<extra_id_73>",
36
+ "<extra_id_72>",
37
+ "<extra_id_71>",
38
+ "<extra_id_70>",
39
+ "<extra_id_69>",
40
+ "<extra_id_68>",
41
+ "<extra_id_67>",
42
+ "<extra_id_66>",
43
+ "<extra_id_65>",
44
+ "<extra_id_64>",
45
+ "<extra_id_63>",
46
+ "<extra_id_62>",
47
+ "<extra_id_61>",
48
+ "<extra_id_60>",
49
+ "<extra_id_59>",
50
+ "<extra_id_58>",
51
+ "<extra_id_57>",
52
+ "<extra_id_56>",
53
+ "<extra_id_55>",
54
+ "<extra_id_54>",
55
+ "<extra_id_53>",
56
+ "<extra_id_52>",
57
+ "<extra_id_51>",
58
+ "<extra_id_50>",
59
+ "<extra_id_49>",
60
+ "<extra_id_48>",
61
+ "<extra_id_47>",
62
+ "<extra_id_46>",
63
+ "<extra_id_45>",
64
+ "<extra_id_44>",
65
+ "<extra_id_43>",
66
+ "<extra_id_42>",
67
+ "<extra_id_41>",
68
+ "<extra_id_40>",
69
+ "<extra_id_39>",
70
+ "<extra_id_38>",
71
+ "<extra_id_37>",
72
+ "<extra_id_36>",
73
+ "<extra_id_35>",
74
+ "<extra_id_34>",
75
+ "<extra_id_33>",
76
+ "<extra_id_32>",
77
+ "<extra_id_31>",
78
+ "<extra_id_30>",
79
+ "<extra_id_29>",
80
+ "<extra_id_28>",
81
+ "<extra_id_27>",
82
+ "<extra_id_26>",
83
+ "<extra_id_25>",
84
+ "<extra_id_24>",
85
+ "<extra_id_23>",
86
+ "<extra_id_22>",
87
+ "<extra_id_21>",
88
+ "<extra_id_20>",
89
+ "<extra_id_19>",
90
+ "<extra_id_18>",
91
+ "<extra_id_17>",
92
+ "<extra_id_16>",
93
+ "<extra_id_15>",
94
+ "<extra_id_14>",
95
+ "<extra_id_13>",
96
+ "<extra_id_12>",
97
+ "<extra_id_11>",
98
+ "<extra_id_10>",
99
+ "<extra_id_9>",
100
+ "<extra_id_8>",
101
+ "<extra_id_7>",
102
+ "<extra_id_6>",
103
+ "<extra_id_5>",
104
+ "<extra_id_4>",
105
+ "<extra_id_3>",
106
+ "<extra_id_2>",
107
+ "<extra_id_1>",
108
+ "<extra_id_0>"
109
+ ],
110
+ "is_local": true,
111
+ "mask_token": "<mask>",
112
+ "max_length": 128,
113
+ "model_max_length": 512,
114
+ "pad_token": "<pad>",
115
+ "sep_token": "</s>",
116
+ "stride": 0,
117
+ "tokenizer_class": "TokenizersBackend",
118
+ "trim_offsets": true,
119
+ "truncation_side": "right",
120
+ "truncation_strategy": "longest_first",
121
+ "unk_token": "<unk>"
122
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b35a5e7fd07656f48c5e4b636b892270a761fa6de6c0e189afe38d1b4b0a8ba8
3
+ size 5393