ludziej commited on
Commit
014b724
1 Parent(s): e240099

End of training

Browse files
README.md CHANGED
@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
13
 
14
  This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
15
  It achieves the following results on the evaluation set:
16
- - Loss: 0.7800
17
 
18
  ## Model description
19
 
@@ -32,119 +32,28 @@ More information needed
32
  ### Training hyperparameters
33
 
34
  The following hyperparameters were used during training:
35
- - learning_rate: 2e-05
36
  - train_batch_size: 512
37
  - eval_batch_size: 512
38
  - seed: 42
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: linear
41
- - num_epochs: 100
42
- - mixed_precision_training: Native AMP
43
 
44
  ### Training results
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:----:|:---------------:|
48
- | No log | 1.0 | 6 | 6.5672 |
49
- | No log | 2.0 | 12 | 5.9069 |
50
- | No log | 3.0 | 18 | 5.4671 |
51
- | No log | 4.0 | 24 | 5.0855 |
52
- | No log | 5.0 | 30 | 4.7085 |
53
- | No log | 6.0 | 36 | 4.3935 |
54
- | No log | 7.0 | 42 | 4.1354 |
55
- | No log | 8.0 | 48 | 3.9330 |
56
- | No log | 9.0 | 54 | 3.7647 |
57
- | No log | 10.0 | 60 | 3.6286 |
58
- | No log | 11.0 | 66 | 3.5173 |
59
- | No log | 12.0 | 72 | 3.4148 |
60
- | No log | 13.0 | 78 | 3.3148 |
61
- | No log | 14.0 | 84 | 3.2218 |
62
- | No log | 15.0 | 90 | 3.1353 |
63
- | No log | 16.0 | 96 | 3.0524 |
64
- | No log | 17.0 | 102 | 2.9751 |
65
- | No log | 18.0 | 108 | 2.9017 |
66
- | No log | 19.0 | 114 | 2.8277 |
67
- | No log | 20.0 | 120 | 2.7534 |
68
- | No log | 21.0 | 126 | 2.6865 |
69
- | No log | 22.0 | 132 | 2.6260 |
70
- | No log | 23.0 | 138 | 2.5597 |
71
- | No log | 24.0 | 144 | 2.4913 |
72
- | No log | 25.0 | 150 | 2.4295 |
73
- | No log | 26.0 | 156 | 2.3668 |
74
- | No log | 27.0 | 162 | 2.3040 |
75
- | No log | 28.0 | 168 | 2.2391 |
76
- | No log | 29.0 | 174 | 2.1745 |
77
- | No log | 30.0 | 180 | 2.1036 |
78
- | No log | 31.0 | 186 | 2.0443 |
79
- | No log | 32.0 | 192 | 1.9919 |
80
- | No log | 33.0 | 198 | 1.9321 |
81
- | No log | 34.0 | 204 | 1.8771 |
82
- | No log | 35.0 | 210 | 1.8386 |
83
- | No log | 36.0 | 216 | 1.7951 |
84
- | No log | 37.0 | 222 | 1.7460 |
85
- | No log | 38.0 | 228 | 1.6974 |
86
- | No log | 39.0 | 234 | 1.6576 |
87
- | No log | 40.0 | 240 | 1.6112 |
88
- | No log | 41.0 | 246 | 1.5811 |
89
- | No log | 42.0 | 252 | 1.5540 |
90
- | No log | 43.0 | 258 | 1.5268 |
91
- | No log | 44.0 | 264 | 1.4873 |
92
- | No log | 45.0 | 270 | 1.4500 |
93
- | No log | 46.0 | 276 | 1.4161 |
94
- | No log | 47.0 | 282 | 1.3738 |
95
- | No log | 48.0 | 288 | 1.3495 |
96
- | No log | 49.0 | 294 | 1.3182 |
97
- | No log | 50.0 | 300 | 1.2899 |
98
- | No log | 51.0 | 306 | 1.2610 |
99
- | No log | 52.0 | 312 | 1.2478 |
100
- | No log | 53.0 | 318 | 1.2238 |
101
- | No log | 54.0 | 324 | 1.2060 |
102
- | No log | 55.0 | 330 | 1.1794 |
103
- | No log | 56.0 | 336 | 1.1774 |
104
- | No log | 57.0 | 342 | 1.1425 |
105
- | No log | 58.0 | 348 | 1.1166 |
106
- | No log | 59.0 | 354 | 1.1044 |
107
- | No log | 60.0 | 360 | 1.0913 |
108
- | No log | 61.0 | 366 | 1.0775 |
109
- | No log | 62.0 | 372 | 1.0694 |
110
- | No log | 63.0 | 378 | 1.0311 |
111
- | No log | 64.0 | 384 | 1.0272 |
112
- | No log | 65.0 | 390 | 1.0249 |
113
- | No log | 66.0 | 396 | 0.9923 |
114
- | No log | 67.0 | 402 | 0.9892 |
115
- | No log | 68.0 | 408 | 0.9762 |
116
- | No log | 69.0 | 414 | 0.9704 |
117
- | No log | 70.0 | 420 | 0.9405 |
118
- | No log | 71.0 | 426 | 0.9394 |
119
- | No log | 72.0 | 432 | 0.9237 |
120
- | No log | 73.0 | 438 | 0.9180 |
121
- | No log | 74.0 | 444 | 0.8926 |
122
- | No log | 75.0 | 450 | 0.9081 |
123
- | No log | 76.0 | 456 | 0.8778 |
124
- | No log | 77.0 | 462 | 0.8785 |
125
- | No log | 78.0 | 468 | 0.8580 |
126
- | No log | 79.0 | 474 | 0.8593 |
127
- | No log | 80.0 | 480 | 0.8553 |
128
- | No log | 81.0 | 486 | 0.8671 |
129
- | No log | 82.0 | 492 | 0.8422 |
130
- | No log | 83.0 | 498 | 0.8403 |
131
- | 2.0956 | 84.0 | 504 | 0.8355 |
132
- | 2.0956 | 85.0 | 510 | 0.8188 |
133
- | 2.0956 | 86.0 | 516 | 0.8149 |
134
- | 2.0956 | 87.0 | 522 | 0.8285 |
135
- | 2.0956 | 88.0 | 528 | 0.8063 |
136
- | 2.0956 | 89.0 | 534 | 0.8166 |
137
- | 2.0956 | 90.0 | 540 | 0.8008 |
138
- | 2.0956 | 91.0 | 546 | 0.8127 |
139
- | 2.0956 | 92.0 | 552 | 0.7921 |
140
- | 2.0956 | 93.0 | 558 | 0.8015 |
141
- | 2.0956 | 94.0 | 564 | 0.7882 |
142
- | 2.0956 | 95.0 | 570 | 0.7844 |
143
- | 2.0956 | 96.0 | 576 | 0.7862 |
144
- | 2.0956 | 97.0 | 582 | 0.7810 |
145
- | 2.0956 | 98.0 | 588 | 0.7808 |
146
- | 2.0956 | 99.0 | 594 | 0.7810 |
147
- | 2.0956 | 100.0 | 600 | 0.7800 |
148
 
149
 
150
  ### Framework versions
 
13
 
14
  This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
15
  It achieves the following results on the evaluation set:
16
+ - Loss: 0.8854
17
 
18
  ## Model description
19
 
 
32
  ### Training hyperparameters
33
 
34
  The following hyperparameters were used during training:
35
+ - learning_rate: 0.001
36
  - train_batch_size: 512
37
  - eval_batch_size: 512
38
  - seed: 42
39
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
40
  - lr_scheduler_type: linear
41
+ - num_epochs: 10
 
42
 
43
  ### Training results
44
 
45
  | Training Loss | Epoch | Step | Validation Loss |
46
  |:-------------:|:-----:|:----:|:---------------:|
47
+ | 3.0459 | 1.0 | 5 | 2.4452 |
48
+ | 2.2781 | 2.0 | 10 | 2.0536 |
49
+ | 1.883 | 3.0 | 15 | 1.6599 |
50
+ | 1.5519 | 4.0 | 20 | 1.4205 |
51
+ | 1.3589 | 5.0 | 25 | 1.2785 |
52
+ | 1.2145 | 6.0 | 30 | 1.1344 |
53
+ | 1.0845 | 7.0 | 35 | 1.0183 |
54
+ | 0.9945 | 8.0 | 40 | 0.9482 |
55
+ | 0.9314 | 9.0 | 45 | 0.8990 |
56
+ | 0.8972 | 10.0 | 50 | 0.8854 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
57
 
58
 
59
  ### Framework versions
config.json CHANGED
@@ -25,13 +25,13 @@
25
  "forced_eos_token_id": null,
26
  "hidden_act": "gelu",
27
  "hidden_dropout_prob": 0.1,
28
- "hidden_size": 768,
29
  "id2label": {
30
  "0": "LABEL_0",
31
  "1": "LABEL_1"
32
  },
33
  "initializer_range": 0.02,
34
- "intermediate_size": 3072,
35
  "is_decoder": true,
36
  "is_encoder_decoder": false,
37
  "label2id": {
@@ -45,10 +45,10 @@
45
  "min_length": 0,
46
  "model_type": "bert",
47
  "no_repeat_ngram_size": 0,
48
- "num_attention_heads": 12,
49
  "num_beam_groups": 1,
50
  "num_beams": 1,
51
- "num_hidden_layers": 12,
52
  "num_return_sequences": 1,
53
  "output_attentions": false,
54
  "output_hidden_states": false,
@@ -78,9 +78,9 @@
78
  "typical_p": 1.0,
79
  "use_bfloat16": false,
80
  "use_cache": true,
81
- "vocab_size": 30522
82
  },
83
- "decoder_start_token_id": 50,
84
  "encoder": {
85
  "_name_or_path": "",
86
  "add_cross_attention": false,
@@ -104,13 +104,13 @@
104
  "forced_eos_token_id": null,
105
  "hidden_act": "gelu",
106
  "hidden_dropout_prob": 0.1,
107
- "hidden_size": 768,
108
  "id2label": {
109
  "0": "LABEL_0",
110
  "1": "LABEL_1"
111
  },
112
  "initializer_range": 0.02,
113
- "intermediate_size": 3072,
114
  "is_decoder": false,
115
  "is_encoder_decoder": false,
116
  "label2id": {
@@ -124,10 +124,10 @@
124
  "min_length": 0,
125
  "model_type": "bert",
126
  "no_repeat_ngram_size": 0,
127
- "num_attention_heads": 12,
128
  "num_beam_groups": 1,
129
  "num_beams": 1,
130
- "num_hidden_layers": 12,
131
  "num_return_sequences": 1,
132
  "output_attentions": false,
133
  "output_hidden_states": false,
@@ -157,11 +157,12 @@
157
  "typical_p": 1.0,
158
  "use_bfloat16": false,
159
  "use_cache": true,
160
- "vocab_size": 30522
161
  },
 
162
  "is_encoder_decoder": true,
163
  "model_type": "encoder-decoder",
164
- "pad_token_id": 49,
165
  "torch_dtype": "float32",
166
  "transformers_version": "4.37.2"
167
  }
 
25
  "forced_eos_token_id": null,
26
  "hidden_act": "gelu",
27
  "hidden_dropout_prob": 0.1,
28
+ "hidden_size": 256,
29
  "id2label": {
30
  "0": "LABEL_0",
31
  "1": "LABEL_1"
32
  },
33
  "initializer_range": 0.02,
34
+ "intermediate_size": 1024,
35
  "is_decoder": true,
36
  "is_encoder_decoder": false,
37
  "label2id": {
 
45
  "min_length": 0,
46
  "model_type": "bert",
47
  "no_repeat_ngram_size": 0,
48
+ "num_attention_heads": 4,
49
  "num_beam_groups": 1,
50
  "num_beams": 1,
51
+ "num_hidden_layers": 4,
52
  "num_return_sequences": 1,
53
  "output_attentions": false,
54
  "output_hidden_states": false,
 
78
  "typical_p": 1.0,
79
  "use_bfloat16": false,
80
  "use_cache": true,
81
+ "vocab_size": 51
82
  },
83
+ "decoder_start_token_id": 49,
84
  "encoder": {
85
  "_name_or_path": "",
86
  "add_cross_attention": false,
 
104
  "forced_eos_token_id": null,
105
  "hidden_act": "gelu",
106
  "hidden_dropout_prob": 0.1,
107
+ "hidden_size": 256,
108
  "id2label": {
109
  "0": "LABEL_0",
110
  "1": "LABEL_1"
111
  },
112
  "initializer_range": 0.02,
113
+ "intermediate_size": 1024,
114
  "is_decoder": false,
115
  "is_encoder_decoder": false,
116
  "label2id": {
 
124
  "min_length": 0,
125
  "model_type": "bert",
126
  "no_repeat_ngram_size": 0,
127
+ "num_attention_heads": 4,
128
  "num_beam_groups": 1,
129
  "num_beams": 1,
130
+ "num_hidden_layers": 4,
131
  "num_return_sequences": 1,
132
  "output_attentions": false,
133
  "output_hidden_states": false,
 
157
  "typical_p": 1.0,
158
  "use_bfloat16": false,
159
  "use_cache": true,
160
+ "vocab_size": 51
161
  },
162
+ "eos_token_id": 50,
163
  "is_encoder_decoder": true,
164
  "model_type": "encoder-decoder",
165
+ "pad_token_id": 48,
166
  "torch_dtype": "float32",
167
  "transformers_version": "4.37.2"
168
  }
generation_config.json CHANGED
@@ -1,6 +1,7 @@
1
  {
2
  "_from_model_config": true,
3
- "decoder_start_token_id": 50,
4
- "pad_token_id": 49,
 
5
  "transformers_version": "4.37.2"
6
  }
 
1
  {
2
  "_from_model_config": true,
3
+ "decoder_start_token_id": 49,
4
+ "eos_token_id": 50,
5
+ "pad_token_id": 48,
6
  "transformers_version": "4.37.2"
7
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fae817cbabf8915ca265feb78d2fad2a1c3568e7be552c9c73e7b1732c4434af
3
- size 989518784
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d68eedd7c2ce4a569ddec467812d03352c8a7b222c55e6dde6b7c022a7ea4859
3
+ size 31203500
runs/Feb28_12-49-12_3897ec21fae5/events.out.tfevents.1709124554.3897ec21fae5.522.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:47ae2b789df2104932122b9ffa05fe48eb5d7ce55098569eb2c03d539761f1ec
3
+ size 10840
runs/Feb28_12-53-45_3897ec21fae5/events.out.tfevents.1709124826.3897ec21fae5.9652.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b85740a246987aa9d85ecec16059f1234089704a567374d5a00b2d22c0c6bd8d
3
+ size 8739
runs/Feb28_12-55-56_3897ec21fae5/events.out.tfevents.1709124958.3897ec21fae5.10174.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e7f99781de3f25bf836d547c6645e62aece21c6afa1f606efa4378d8464f234
3
+ size 12867
special_tokens_map.json CHANGED
@@ -6,6 +6,13 @@
6
  "rstrip": false,
7
  "single_word": false
8
  },
 
 
 
 
 
 
 
9
  "pad_token": {
10
  "content": "[PAD]",
11
  "lstrip": false,
 
6
  "rstrip": false,
7
  "single_word": false
8
  },
9
+ "eos_token": {
10
+ "content": "[EOS]",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
  "pad_token": {
17
  "content": "[PAD]",
18
  "lstrip": false,
tokenizer.json CHANGED
@@ -4,7 +4,7 @@
4
  "padding": null,
5
  "added_tokens": [
6
  {
7
- "id": 49,
8
  "content": "[PAD]",
9
  "single_word": false,
10
  "lstrip": false,
@@ -13,13 +13,22 @@
13
  "special": true
14
  },
15
  {
16
- "id": 50,
17
  "content": "[CLS]",
18
  "single_word": false,
19
  "lstrip": false,
20
  "rstrip": false,
21
  "normalized": false,
22
  "special": true
 
 
 
 
 
 
 
 
 
23
  }
24
  ],
25
  "normalizer": null,
@@ -50,80 +59,78 @@
50
  "8": 10,
51
  "9": 11,
52
  "=": 12,
53
- "99": 13,
54
- "10": 14,
55
- "98": 15,
56
- "11": 16,
57
- "97": 17,
58
- "12": 18,
59
  "96": 19,
60
  "13": 20,
61
- "95": 21,
62
- "14": 22,
63
- "94": 23,
64
- "15": 24,
65
- "93": 25,
66
- "16": 26,
67
  "17": 27,
68
  "92": 28,
69
- "91": 29,
70
- "18": 30,
71
- "90": 31,
72
- "19": 32,
73
- "89": 33,
74
- "20": 34,
75
- "21": 35,
76
- "88": 36,
77
  "22": 37,
78
  "87": 38,
79
  "23": 39,
80
  "86": 40,
81
- "85": 41,
82
- "24": 42,
83
- "25": 43,
84
- "84": 44,
85
  "26": 45,
86
  "83": 46,
87
- "27": 47,
88
- "82": 48
89
  },
90
  "merges": [
91
- "9 9",
92
  "1 0",
93
- "9 8",
94
  "1 1",
95
- "9 7",
96
  "1 2",
 
97
  "9 6",
98
  "1 3",
99
- "9 5",
100
  "1 4",
101
- "9 4",
102
  "1 5",
103
- "9 3",
104
  "1 6",
 
105
  "1 7",
106
  "9 2",
107
- "9 1",
108
  "1 8",
109
- "9 0",
110
  "1 9",
111
- "8 9",
112
  "2 0",
113
- "2 1",
114
  "8 8",
 
115
  "2 2",
116
  "8 7",
117
  "2 3",
118
  "8 6",
119
- "8 5",
120
  "2 4",
121
- "2 5",
122
  "8 4",
 
123
  "2 6",
124
  "8 3",
125
- "2 7",
126
- "8 2"
127
  ]
128
  }
129
  }
 
4
  "padding": null,
5
  "added_tokens": [
6
  {
7
+ "id": 48,
8
  "content": "[PAD]",
9
  "single_word": false,
10
  "lstrip": false,
 
13
  "special": true
14
  },
15
  {
16
+ "id": 49,
17
  "content": "[CLS]",
18
  "single_word": false,
19
  "lstrip": false,
20
  "rstrip": false,
21
  "normalized": false,
22
  "special": true
23
+ },
24
+ {
25
+ "id": 50,
26
+ "content": "[EOS]",
27
+ "single_word": false,
28
+ "lstrip": false,
29
+ "rstrip": false,
30
+ "normalized": false,
31
+ "special": true
32
  }
33
  ],
34
  "normalizer": null,
 
59
  "8": 10,
60
  "9": 11,
61
  "=": 12,
62
+ "10": 13,
63
+ "99": 14,
64
+ "11": 15,
65
+ "98": 16,
66
+ "12": 17,
67
+ "97": 18,
68
  "96": 19,
69
  "13": 20,
70
+ "14": 21,
71
+ "95": 22,
72
+ "15": 23,
73
+ "94": 24,
74
+ "16": 25,
75
+ "93": 26,
76
  "17": 27,
77
  "92": 28,
78
+ "18": 29,
79
+ "91": 30,
80
+ "19": 31,
81
+ "90": 32,
82
+ "20": 33,
83
+ "89": 34,
84
+ "88": 35,
85
+ "21": 36,
86
  "22": 37,
87
  "87": 38,
88
  "23": 39,
89
  "86": 40,
90
+ "24": 41,
91
+ "85": 42,
92
+ "84": 43,
93
+ "25": 44,
94
  "26": 45,
95
  "83": 46,
96
+ "27": 47
 
97
  },
98
  "merges": [
 
99
  "1 0",
100
+ "9 9",
101
  "1 1",
102
+ "9 8",
103
  "1 2",
104
+ "9 7",
105
  "9 6",
106
  "1 3",
 
107
  "1 4",
108
+ "9 5",
109
  "1 5",
110
+ "9 4",
111
  "1 6",
112
+ "9 3",
113
  "1 7",
114
  "9 2",
 
115
  "1 8",
116
+ "9 1",
117
  "1 9",
118
+ "9 0",
119
  "2 0",
120
+ "8 9",
121
  "8 8",
122
+ "2 1",
123
  "2 2",
124
  "8 7",
125
  "2 3",
126
  "8 6",
 
127
  "2 4",
128
+ "8 5",
129
  "8 4",
130
+ "2 5",
131
  "2 6",
132
  "8 3",
133
+ "2 7"
 
134
  ]
135
  }
136
  }
tokenizer_config.json CHANGED
@@ -1,6 +1,6 @@
1
  {
2
  "added_tokens_decoder": {
3
- "49": {
4
  "content": "[PAD]",
5
  "lstrip": false,
6
  "normalized": false,
@@ -8,17 +8,26 @@
8
  "single_word": false,
9
  "special": true
10
  },
11
- "50": {
12
  "content": "[CLS]",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
16
  "single_word": false,
17
  "special": true
 
 
 
 
 
 
 
 
18
  }
19
  },
20
  "clean_up_tokenization_spaces": true,
21
  "cls_token": "[CLS]",
 
22
  "model_max_length": 1000000000000000019884624838656,
23
  "pad_token": "[PAD]",
24
  "tokenizer_class": "PreTrainedTokenizerFast"
 
1
  {
2
  "added_tokens_decoder": {
3
+ "48": {
4
  "content": "[PAD]",
5
  "lstrip": false,
6
  "normalized": false,
 
8
  "single_word": false,
9
  "special": true
10
  },
11
+ "49": {
12
  "content": "[CLS]",
13
  "lstrip": false,
14
  "normalized": false,
15
  "rstrip": false,
16
  "single_word": false,
17
  "special": true
18
+ },
19
+ "50": {
20
+ "content": "[EOS]",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
  }
27
  },
28
  "clean_up_tokenization_spaces": true,
29
  "cls_token": "[CLS]",
30
+ "eos_token": "[EOS]",
31
  "model_max_length": 1000000000000000019884624838656,
32
  "pad_token": "[PAD]",
33
  "tokenizer_class": "PreTrainedTokenizerFast"
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:12b3e8c762fa264d4d0652b207fff70f26ff840984ec4601c4b064c0086cc0a9
3
  size 4920
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c22510d433bcd6c8ab9f638615a22414f615f7dc618aab32a485345ef2763c73
3
  size 4920