pierreguillou commited on
Commit
ed2f289
1 Parent(s): ea9a810

Training in progress, step 6400

Browse files
checkpoint-6400/config.json ADDED
@@ -0,0 +1,161 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "microsoft/layoutxlm-base",
3
+ "architectures": [
4
+ "LayoutLMv2ForTokenClassification"
5
+ ],
6
+ "attention_probs_dropout_prob": 0.1,
7
+ "bos_token_id": 0,
8
+ "convert_sync_batchnorm": true,
9
+ "coordinate_size": 128,
10
+ "detectron2_config_args": {
11
+ "MODEL.ANCHOR_GENERATOR.SIZES": [
12
+ [
13
+ 32
14
+ ],
15
+ [
16
+ 64
17
+ ],
18
+ [
19
+ 128
20
+ ],
21
+ [
22
+ 256
23
+ ],
24
+ [
25
+ 512
26
+ ]
27
+ ],
28
+ "MODEL.BACKBONE.NAME": "build_resnet_fpn_backbone",
29
+ "MODEL.FPN.IN_FEATURES": [
30
+ "res2",
31
+ "res3",
32
+ "res4",
33
+ "res5"
34
+ ],
35
+ "MODEL.MASK_ON": true,
36
+ "MODEL.PIXEL_STD": [
37
+ 57.375,
38
+ 57.12,
39
+ 58.395
40
+ ],
41
+ "MODEL.POST_NMS_TOPK_TEST": 1000,
42
+ "MODEL.RESNETS.ASPECT_RATIOS": [
43
+ [
44
+ 0.5,
45
+ 1.0,
46
+ 2.0
47
+ ]
48
+ ],
49
+ "MODEL.RESNETS.DEPTH": 101,
50
+ "MODEL.RESNETS.NUM_GROUPS": 32,
51
+ "MODEL.RESNETS.OUT_FEATURES": [
52
+ "res2",
53
+ "res3",
54
+ "res4",
55
+ "res5"
56
+ ],
57
+ "MODEL.RESNETS.SIZES": [
58
+ [
59
+ 32
60
+ ],
61
+ [
62
+ 64
63
+ ],
64
+ [
65
+ 128
66
+ ],
67
+ [
68
+ 256
69
+ ],
70
+ [
71
+ 512
72
+ ]
73
+ ],
74
+ "MODEL.RESNETS.STRIDE_IN_1X1": false,
75
+ "MODEL.RESNETS.WIDTH_PER_GROUP": 8,
76
+ "MODEL.ROI_BOX_HEAD.NAME": "FastRCNNConvFCHead",
77
+ "MODEL.ROI_BOX_HEAD.NUM_FC": 2,
78
+ "MODEL.ROI_BOX_HEAD.POOLER_RESOLUTION": 14,
79
+ "MODEL.ROI_HEADS.IN_FEATURES": [
80
+ "p2",
81
+ "p3",
82
+ "p4",
83
+ "p5"
84
+ ],
85
+ "MODEL.ROI_HEADS.NAME": "StandardROIHeads",
86
+ "MODEL.ROI_HEADS.NUM_CLASSES": 5,
87
+ "MODEL.ROI_MASK_HEAD.NAME": "MaskRCNNConvUpsampleHead",
88
+ "MODEL.ROI_MASK_HEAD.NUM_CONV": 4,
89
+ "MODEL.ROI_MASK_HEAD.POOLER_RESOLUTION": 7,
90
+ "MODEL.RPN.IN_FEATURES": [
91
+ "p2",
92
+ "p3",
93
+ "p4",
94
+ "p5",
95
+ "p6"
96
+ ],
97
+ "MODEL.RPN.POST_NMS_TOPK_TRAIN": 1000,
98
+ "MODEL.RPN.PRE_NMS_TOPK_TEST": 1000,
99
+ "MODEL.RPN.PRE_NMS_TOPK_TRAIN": 2000
100
+ },
101
+ "eos_token_id": 2,
102
+ "fast_qkv": false,
103
+ "gradient_checkpointing": false,
104
+ "has_relative_attention_bias": false,
105
+ "has_spatial_attention_bias": false,
106
+ "has_visual_segment_embedding": true,
107
+ "hidden_act": "gelu",
108
+ "hidden_dropout_prob": 0.1,
109
+ "hidden_size": 768,
110
+ "id2label": {
111
+ "0": "Caption",
112
+ "1": "Footnote",
113
+ "2": "Formula",
114
+ "3": "List-item",
115
+ "4": "Page-footer",
116
+ "5": "Page-header",
117
+ "6": "Picture",
118
+ "7": "Section-header",
119
+ "8": "Table",
120
+ "9": "Text",
121
+ "10": "Title"
122
+ },
123
+ "image_feature_pool_shape": [
124
+ 7,
125
+ 7,
126
+ 256
127
+ ],
128
+ "initializer_range": 0.02,
129
+ "intermediate_size": 3072,
130
+ "label2id": {
131
+ "Caption": 0,
132
+ "Footnote": 1,
133
+ "Formula": 2,
134
+ "List-item": 3,
135
+ "Page-footer": 4,
136
+ "Page-header": 5,
137
+ "Picture": 6,
138
+ "Section-header": 7,
139
+ "Table": 8,
140
+ "Text": 9,
141
+ "Title": 10
142
+ },
143
+ "layer_norm_eps": 1e-05,
144
+ "max_2d_position_embeddings": 1024,
145
+ "max_position_embeddings": 514,
146
+ "max_rel_2d_pos": 256,
147
+ "max_rel_pos": 128,
148
+ "model_type": "layoutlmv2",
149
+ "num_attention_heads": 12,
150
+ "num_hidden_layers": 12,
151
+ "output_past": true,
152
+ "pad_token_id": 1,
153
+ "rel_2d_pos_bins": 64,
154
+ "rel_pos_bins": 32,
155
+ "shape_size": 128,
156
+ "tokenizer_class": "LayoutXLMTokenizer",
157
+ "torch_dtype": "float32",
158
+ "transformers_version": "4.27.3",
159
+ "type_vocab_size": 1,
160
+ "vocab_size": 250002
161
+ }
checkpoint-6400/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2a9be5bb8c43d23118edb934f424c908173b167f7c17e762b94d88081cdff147
3
+ size 2927175013
checkpoint-6400/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3c5b87531e31703b8be687e53ab9cabd0de73011638a31ed129e72c1b0b6baf
3
+ size 1476527911
checkpoint-6400/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e38861dae05a4731c6abb772b7018dca4557b45be036c759381f98f7f2c97820
3
+ size 14503
checkpoint-6400/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ce30ac03f175cabec41f9201ef6f56bcd0012c23f2e1f5f20c12f4886a8ebe8
3
+ size 559
checkpoint-6400/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:778e6ce48e52036842ebf13ab098ed17c0f3d7b9dc15ff5bebd335f0702788fc
3
+ size 623
checkpoint-6400/sentencepiece.bpe.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cfc8146abe2a0488e9e2a0c56de7952f7c11ab059eca145a0a727afce0db2865
3
+ size 5069051
checkpoint-6400/special_tokens_map.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "eos_token": "</s>",
5
+ "mask_token": {
6
+ "content": "<mask>",
7
+ "lstrip": true,
8
+ "normalized": true,
9
+ "rstrip": false,
10
+ "single_word": false
11
+ },
12
+ "pad_token": "<pad>",
13
+ "sep_token": "</s>",
14
+ "unk_token": "<unk>"
15
+ }
checkpoint-6400/tokenizer_config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "cls_token": "<s>",
4
+ "cls_token_box": [
5
+ 0,
6
+ 0,
7
+ 0,
8
+ 0
9
+ ],
10
+ "eos_token": "</s>",
11
+ "mask_token": {
12
+ "__type": "AddedToken",
13
+ "content": "<mask>",
14
+ "lstrip": true,
15
+ "normalized": true,
16
+ "rstrip": false,
17
+ "single_word": false
18
+ },
19
+ "model_max_length": 1000000000000000019884624838656,
20
+ "only_label_first_subword": true,
21
+ "pad_token": "<pad>",
22
+ "pad_token_box": [
23
+ 0,
24
+ 0,
25
+ 0,
26
+ 0
27
+ ],
28
+ "pad_token_label": -100,
29
+ "sep_token": "</s>",
30
+ "sep_token_box": [
31
+ 1000,
32
+ 1000,
33
+ 1000,
34
+ 1000
35
+ ],
36
+ "sp_model_kwargs": {},
37
+ "special_tokens_map_file": null,
38
+ "tokenizer_class": "LayoutXLMTokenizer",
39
+ "unk_token": "<unk>"
40
+ }
checkpoint-6400/trainer_state.json ADDED
@@ -0,0 +1,472 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": 0.7893899204244033,
3
+ "best_model_checkpoint": "DocLayNet/layout-xlm-base-finetuned-DocLayNet-base_paragraphs_ml512-v6/checkpoint-6400",
4
+ "epoch": 3.411513859275053,
5
+ "global_step": 6400,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.11,
12
+ "eval_accuracy": 0.8841533429944805,
13
+ "eval_f1": 0.10664094791953704,
14
+ "eval_loss": 0.4427741765975952,
15
+ "eval_precision": 0.11536168521462639,
16
+ "eval_recall": 0.09914602903501281,
17
+ "eval_runtime": 348.7843,
18
+ "eval_samples_per_second": 4.607,
19
+ "eval_steps_per_second": 0.29,
20
+ "step": 200
21
+ },
22
+ {
23
+ "epoch": 0.21,
24
+ "eval_accuracy": 0.9242655812394248,
25
+ "eval_f1": 0.44395873404441333,
26
+ "eval_loss": 0.30403900146484375,
27
+ "eval_precision": 0.4547734193086154,
28
+ "eval_recall": 0.4336464560204953,
29
+ "eval_runtime": 350.1211,
30
+ "eval_samples_per_second": 4.59,
31
+ "eval_steps_per_second": 0.288,
32
+ "step": 400
33
+ },
34
+ {
35
+ "epoch": 0.27,
36
+ "learning_rate": 1.3209054593874834e-05,
37
+ "loss": 0.7241,
38
+ "step": 500
39
+ },
40
+ {
41
+ "epoch": 0.32,
42
+ "eval_accuracy": 0.9359434268564053,
43
+ "eval_f1": 0.5543897040741045,
44
+ "eval_loss": 0.22647875547409058,
45
+ "eval_precision": 0.5330233291298865,
46
+ "eval_recall": 0.5775405636208369,
47
+ "eval_runtime": 342.3688,
48
+ "eval_samples_per_second": 4.694,
49
+ "eval_steps_per_second": 0.295,
50
+ "step": 600
51
+ },
52
+ {
53
+ "epoch": 0.43,
54
+ "eval_accuracy": 0.947861481323102,
55
+ "eval_f1": 0.6015281512784394,
56
+ "eval_loss": 0.2140066921710968,
57
+ "eval_precision": 0.6013484680378937,
58
+ "eval_recall": 0.6017079419299743,
59
+ "eval_runtime": 342.7133,
60
+ "eval_samples_per_second": 4.689,
61
+ "eval_steps_per_second": 0.295,
62
+ "step": 800
63
+ },
64
+ {
65
+ "epoch": 0.53,
66
+ "learning_rate": 1.927439656448986e-05,
67
+ "loss": 0.2343,
68
+ "step": 1000
69
+ },
70
+ {
71
+ "epoch": 0.53,
72
+ "eval_accuracy": 0.9402091136821377,
73
+ "eval_f1": 0.6132131856006622,
74
+ "eval_loss": 0.285193532705307,
75
+ "eval_precision": 0.6641768748132656,
76
+ "eval_recall": 0.5695132365499573,
77
+ "eval_runtime": 341.4863,
78
+ "eval_samples_per_second": 4.706,
79
+ "eval_steps_per_second": 0.296,
80
+ "step": 1000
81
+ },
82
+ {
83
+ "epoch": 0.64,
84
+ "eval_accuracy": 0.9539986854504707,
85
+ "eval_f1": 0.6604133950171894,
86
+ "eval_loss": 0.16944007575511932,
87
+ "eval_precision": 0.6564846848367226,
88
+ "eval_recall": 0.6643894107600341,
89
+ "eval_runtime": 342.4627,
90
+ "eval_samples_per_second": 4.692,
91
+ "eval_steps_per_second": 0.295,
92
+ "step": 1200
93
+ },
94
+ {
95
+ "epoch": 0.75,
96
+ "eval_accuracy": 0.9354234142916347,
97
+ "eval_f1": 0.6198065842234727,
98
+ "eval_loss": 0.230770543217659,
99
+ "eval_precision": 0.5118828964212166,
100
+ "eval_recall": 0.785397096498719,
101
+ "eval_runtime": 332.1227,
102
+ "eval_samples_per_second": 4.839,
103
+ "eval_steps_per_second": 0.304,
104
+ "step": 1400
105
+ },
106
+ {
107
+ "epoch": 0.8,
108
+ "learning_rate": 1.7793573226714055e-05,
109
+ "loss": 0.1913,
110
+ "step": 1500
111
+ },
112
+ {
113
+ "epoch": 0.85,
114
+ "eval_accuracy": 0.9593994250825553,
115
+ "eval_f1": 0.6589563286454478,
116
+ "eval_loss": 0.1601293832063675,
117
+ "eval_precision": 0.7189582071471835,
118
+ "eval_recall": 0.6081981212638771,
119
+ "eval_runtime": 327.08,
120
+ "eval_samples_per_second": 4.913,
121
+ "eval_steps_per_second": 0.309,
122
+ "step": 1600
123
+ },
124
+ {
125
+ "epoch": 0.96,
126
+ "eval_accuracy": 0.9540725958657681,
127
+ "eval_f1": 0.6596589238459277,
128
+ "eval_loss": 0.16707618534564972,
129
+ "eval_precision": 0.5790424570912376,
130
+ "eval_recall": 0.7663535439795047,
131
+ "eval_runtime": 324.9043,
132
+ "eval_samples_per_second": 4.946,
133
+ "eval_steps_per_second": 0.311,
134
+ "step": 1800
135
+ },
136
+ {
137
+ "epoch": 1.07,
138
+ "learning_rate": 1.6315711535613804e-05,
139
+ "loss": 0.1346,
140
+ "step": 2000
141
+ },
142
+ {
143
+ "epoch": 1.07,
144
+ "eval_accuracy": 0.9611943923112051,
145
+ "eval_f1": 0.6985545993817363,
146
+ "eval_loss": 0.15802155435085297,
147
+ "eval_precision": 0.6837585868498528,
148
+ "eval_recall": 0.71400512382579,
149
+ "eval_runtime": 326.3436,
150
+ "eval_samples_per_second": 4.924,
151
+ "eval_steps_per_second": 0.309,
152
+ "step": 2000
153
+ },
154
+ {
155
+ "epoch": 1.17,
156
+ "eval_accuracy": 0.9597082650321906,
157
+ "eval_f1": 0.6896523517382412,
158
+ "eval_loss": 0.14230071008205414,
159
+ "eval_precision": 0.661773940345369,
160
+ "eval_recall": 0.7199829205807002,
161
+ "eval_runtime": 352.3832,
162
+ "eval_samples_per_second": 4.56,
163
+ "eval_steps_per_second": 0.287,
164
+ "step": 2200
165
+ },
166
+ {
167
+ "epoch": 1.28,
168
+ "eval_accuracy": 0.9662862919936542,
169
+ "eval_f1": 0.6980116745713244,
170
+ "eval_loss": 0.15801101922988892,
171
+ "eval_precision": 0.7489724016441573,
172
+ "eval_recall": 0.6535439795046969,
173
+ "eval_runtime": 346.3949,
174
+ "eval_samples_per_second": 4.639,
175
+ "eval_steps_per_second": 0.292,
176
+ "step": 2400
177
+ },
178
+ {
179
+ "epoch": 1.33,
180
+ "learning_rate": 1.483784984451355e-05,
181
+ "loss": 0.098,
182
+ "step": 2500
183
+ },
184
+ {
185
+ "epoch": 1.39,
186
+ "eval_accuracy": 0.9615850616492053,
187
+ "eval_f1": 0.6800371073905552,
188
+ "eval_loss": 0.1394117921590805,
189
+ "eval_precision": 0.7044019401482566,
190
+ "eval_recall": 0.6573014517506405,
191
+ "eval_runtime": 340.6651,
192
+ "eval_samples_per_second": 4.717,
193
+ "eval_steps_per_second": 0.296,
194
+ "step": 2600
195
+ },
196
+ {
197
+ "epoch": 1.49,
198
+ "eval_accuracy": 0.9685537579486693,
199
+ "eval_f1": 0.725116373203805,
200
+ "eval_loss": 0.1756211668252945,
201
+ "eval_precision": 0.6892651019622932,
202
+ "eval_recall": 0.7649017933390265,
203
+ "eval_runtime": 336.4339,
204
+ "eval_samples_per_second": 4.777,
205
+ "eval_steps_per_second": 0.3,
206
+ "step": 2800
207
+ },
208
+ {
209
+ "epoch": 1.6,
210
+ "learning_rate": 1.33599881534133e-05,
211
+ "loss": 0.0999,
212
+ "step": 3000
213
+ },
214
+ {
215
+ "epoch": 1.6,
216
+ "eval_accuracy": 0.9636387153313958,
217
+ "eval_f1": 0.69845832244578,
218
+ "eval_loss": 0.154231458902359,
219
+ "eval_precision": 0.7126733025239957,
220
+ "eval_recall": 0.684799316823228,
221
+ "eval_runtime": 338.3198,
222
+ "eval_samples_per_second": 4.75,
223
+ "eval_steps_per_second": 0.299,
224
+ "step": 3000
225
+ },
226
+ {
227
+ "epoch": 1.71,
228
+ "eval_accuracy": 0.9670465134081412,
229
+ "eval_f1": 0.7096949152542371,
230
+ "eval_loss": 0.11871206015348434,
231
+ "eval_precision": 0.7538166106577052,
232
+ "eval_recall": 0.6704526046114432,
233
+ "eval_runtime": 330.1098,
234
+ "eval_samples_per_second": 4.868,
235
+ "eval_steps_per_second": 0.306,
236
+ "step": 3200
237
+ },
238
+ {
239
+ "epoch": 1.81,
240
+ "eval_accuracy": 0.9585151397566763,
241
+ "eval_f1": 0.7426724326213252,
242
+ "eval_loss": 0.17930778861045837,
243
+ "eval_precision": 0.7601716891710633,
244
+ "eval_recall": 0.7259607173356106,
245
+ "eval_runtime": 334.0583,
246
+ "eval_samples_per_second": 4.811,
247
+ "eval_steps_per_second": 0.302,
248
+ "step": 3400
249
+ },
250
+ {
251
+ "epoch": 1.87,
252
+ "learning_rate": 1.1879164815637496e-05,
253
+ "loss": 0.0972,
254
+ "step": 3500
255
+ },
256
+ {
257
+ "epoch": 1.92,
258
+ "eval_accuracy": 0.9621024345562867,
259
+ "eval_f1": 0.7188762229602368,
260
+ "eval_loss": 0.18358349800109863,
261
+ "eval_precision": 0.7575671585319712,
262
+ "eval_recall": 0.6839453458582408,
263
+ "eval_runtime": 323.7426,
264
+ "eval_samples_per_second": 4.964,
265
+ "eval_steps_per_second": 0.312,
266
+ "step": 3600
267
+ },
268
+ {
269
+ "epoch": 2.03,
270
+ "eval_accuracy": 0.9641745658423015,
271
+ "eval_f1": 0.7188527825286147,
272
+ "eval_loss": 0.1464972347021103,
273
+ "eval_precision": 0.7388443162354638,
274
+ "eval_recall": 0.6999146029035013,
275
+ "eval_runtime": 324.6992,
276
+ "eval_samples_per_second": 4.949,
277
+ "eval_steps_per_second": 0.311,
278
+ "step": 3800
279
+ },
280
+ {
281
+ "epoch": 2.13,
282
+ "learning_rate": 1.0398341477861691e-05,
283
+ "loss": 0.0662,
284
+ "step": 4000
285
+ },
286
+ {
287
+ "epoch": 2.13,
288
+ "eval_accuracy": 0.9690632118826831,
289
+ "eval_f1": 0.7450159228722244,
290
+ "eval_loss": 0.1409263014793396,
291
+ "eval_precision": 0.7615267992508695,
292
+ "eval_recall": 0.7292058070025619,
293
+ "eval_runtime": 314.9474,
294
+ "eval_samples_per_second": 5.102,
295
+ "eval_steps_per_second": 0.321,
296
+ "step": 4000
297
+ },
298
+ {
299
+ "epoch": 2.24,
300
+ "eval_accuracy": 0.9615428271261782,
301
+ "eval_f1": 0.7431866723622383,
302
+ "eval_loss": 0.17204762995243073,
303
+ "eval_precision": 0.7435042735042735,
304
+ "eval_recall": 0.7428693424423569,
305
+ "eval_runtime": 307.7666,
306
+ "eval_samples_per_second": 5.221,
307
+ "eval_steps_per_second": 0.328,
308
+ "step": 4200
309
+ },
310
+ {
311
+ "epoch": 2.35,
312
+ "eval_accuracy": 0.9666637630432086,
313
+ "eval_f1": 0.7338054481470216,
314
+ "eval_loss": 0.14404301345348358,
315
+ "eval_precision": 0.7468824621915627,
316
+ "eval_recall": 0.7211784799316823,
317
+ "eval_runtime": 310.6556,
318
+ "eval_samples_per_second": 5.173,
319
+ "eval_steps_per_second": 0.325,
320
+ "step": 4400
321
+ },
322
+ {
323
+ "epoch": 2.4,
324
+ "learning_rate": 8.917518140085888e-06,
325
+ "loss": 0.0581,
326
+ "step": 4500
327
+ },
328
+ {
329
+ "epoch": 2.45,
330
+ "eval_accuracy": 0.9657451621673702,
331
+ "eval_f1": 0.7135284422468708,
332
+ "eval_loss": 0.19275733828544617,
333
+ "eval_precision": 0.7457863860694665,
334
+ "eval_recall": 0.6839453458582408,
335
+ "eval_runtime": 303.8599,
336
+ "eval_samples_per_second": 5.289,
337
+ "eval_steps_per_second": 0.332,
338
+ "step": 4600
339
+ },
340
+ {
341
+ "epoch": 2.56,
342
+ "eval_accuracy": 0.969187275794075,
343
+ "eval_f1": 0.7378380713730234,
344
+ "eval_loss": 0.16446392238140106,
345
+ "eval_precision": 0.7466771598461001,
346
+ "eval_recall": 0.7292058070025619,
347
+ "eval_runtime": 304.6717,
348
+ "eval_samples_per_second": 5.275,
349
+ "eval_steps_per_second": 0.332,
350
+ "step": 4800
351
+ },
352
+ {
353
+ "epoch": 2.67,
354
+ "learning_rate": 7.436694802310085e-06,
355
+ "loss": 0.0538,
356
+ "step": 5000
357
+ },
358
+ {
359
+ "epoch": 2.67,
360
+ "eval_accuracy": 0.9655762240752619,
361
+ "eval_f1": 0.7619294158253516,
362
+ "eval_loss": 0.15173019468784332,
363
+ "eval_precision": 0.7699686083013603,
364
+ "eval_recall": 0.7540563620836892,
365
+ "eval_runtime": 301.8453,
366
+ "eval_samples_per_second": 5.324,
367
+ "eval_steps_per_second": 0.335,
368
+ "step": 5000
369
+ },
370
+ {
371
+ "epoch": 2.77,
372
+ "eval_accuracy": 0.9683900991719394,
373
+ "eval_f1": 0.7727910873601738,
374
+ "eval_loss": 0.16763265430927277,
375
+ "eval_precision": 0.8226786230835985,
376
+ "eval_recall": 0.7286080273270709,
377
+ "eval_runtime": 303.0486,
378
+ "eval_samples_per_second": 5.303,
379
+ "eval_steps_per_second": 0.333,
380
+ "step": 5200
381
+ },
382
+ {
383
+ "epoch": 2.88,
384
+ "eval_accuracy": 0.9724974065363203,
385
+ "eval_f1": 0.7607919220424181,
386
+ "eval_loss": 0.12770096957683563,
387
+ "eval_precision": 0.7864891968274227,
388
+ "eval_recall": 0.7367207514944492,
389
+ "eval_runtime": 298.6206,
390
+ "eval_samples_per_second": 5.381,
391
+ "eval_steps_per_second": 0.338,
392
+ "step": 5400
393
+ },
394
+ {
395
+ "epoch": 2.93,
396
+ "learning_rate": 5.955871464534282e-06,
397
+ "loss": 0.0432,
398
+ "step": 5500
399
+ },
400
+ {
401
+ "epoch": 2.99,
402
+ "eval_accuracy": 0.9692585465516832,
403
+ "eval_f1": 0.7784317119612255,
404
+ "eval_loss": 0.15319885313510895,
405
+ "eval_precision": 0.78908580452711,
406
+ "eval_recall": 0.7680614859094791,
407
+ "eval_runtime": 299.3727,
408
+ "eval_samples_per_second": 5.368,
409
+ "eval_steps_per_second": 0.337,
410
+ "step": 5600
411
+ },
412
+ {
413
+ "epoch": 3.09,
414
+ "eval_accuracy": 0.9691767171633182,
415
+ "eval_f1": 0.7783425414364641,
416
+ "eval_loss": 0.1701451987028122,
417
+ "eval_precision": 0.8066880439761795,
418
+ "eval_recall": 0.7519214346712212,
419
+ "eval_runtime": 330.9586,
420
+ "eval_samples_per_second": 4.856,
421
+ "eval_steps_per_second": 0.305,
422
+ "step": 5800
423
+ },
424
+ {
425
+ "epoch": 3.2,
426
+ "learning_rate": 4.475048126758478e-06,
427
+ "loss": 0.0272,
428
+ "step": 6000
429
+ },
430
+ {
431
+ "epoch": 3.2,
432
+ "eval_accuracy": 0.973239150346983,
433
+ "eval_f1": 0.7797986932721173,
434
+ "eval_loss": 0.11586667597293854,
435
+ "eval_precision": 0.8071650520928533,
436
+ "eval_recall": 0.7542271562766866,
437
+ "eval_runtime": 326.8152,
438
+ "eval_samples_per_second": 4.917,
439
+ "eval_steps_per_second": 0.309,
440
+ "step": 6000
441
+ },
442
+ {
443
+ "epoch": 3.3,
444
+ "eval_accuracy": 0.9719905922599957,
445
+ "eval_f1": 0.7796919071382077,
446
+ "eval_loss": 0.18351919949054718,
447
+ "eval_precision": 0.7925893250992501,
448
+ "eval_recall": 0.7672075149444919,
449
+ "eval_runtime": 329.8149,
450
+ "eval_samples_per_second": 4.872,
451
+ "eval_steps_per_second": 0.306,
452
+ "step": 6200
453
+ },
454
+ {
455
+ "epoch": 3.41,
456
+ "eval_accuracy": 0.9730095001280235,
457
+ "eval_f1": 0.7893899204244033,
458
+ "eval_loss": 0.14806130528450012,
459
+ "eval_precision": 0.8183318056828598,
460
+ "eval_recall": 0.7624252775405637,
461
+ "eval_runtime": 319.7431,
462
+ "eval_samples_per_second": 5.026,
463
+ "eval_steps_per_second": 0.316,
464
+ "step": 6400
465
+ }
466
+ ],
467
+ "max_steps": 7504,
468
+ "num_train_epochs": 4,
469
+ "total_flos": 2.76412871737344e+16,
470
+ "trial_name": null,
471
+ "trial_params": null
472
+ }
checkpoint-6400/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa15259c372902c8706d91a1e90c5ae1b05ee943ad6d35334cb341cb0ae39556
3
+ size 3759
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8ff92c6b665e6a13d266d88a549d4d1bcba52577bb78f626525e18384a1e2e39
3
  size 1476527911
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b3c5b87531e31703b8be687e53ab9cabd0de73011638a31ed129e72c1b0b6baf
3
  size 1476527911
runs/Mar26_08-28-55_341d465657da/events.out.tfevents.1679821115.341d465657da.354.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c3500ea441429b1cfe039c551a77ef6b878a87fbd02616b72ceb986fabb117ed
3
- size 8304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8927b8273ff549e7669e574d6dcddd6d880aead11edb46d72d5c11f5c185dfbe
3
+ size 8776