juliamendelsohn
commited on
Commit
·
9caec0f
1
Parent(s):
68d5ded
commit from juliame
Browse files- config.json +48 -0
- eval_results.txt +5 -0
- merges.txt +0 -0
- model_args.json +1 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
- training_progress_scores.csv +42 -0
- vocab.json +0 -0
config.json
ADDED
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"RobertaForMultiLabelSequenceClassification"
|
4 |
+
],
|
5 |
+
"attention_probs_dropout_prob": 0.1,
|
6 |
+
"bos_token_id": 0,
|
7 |
+
"eos_token_id": 2,
|
8 |
+
"gradient_checkpointing": false,
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 768,
|
12 |
+
"id2label": {
|
13 |
+
"0": "LABEL_0",
|
14 |
+
"1": "LABEL_1",
|
15 |
+
"2": "LABEL_2",
|
16 |
+
"3": "LABEL_3",
|
17 |
+
"4": "LABEL_4",
|
18 |
+
"5": "LABEL_5",
|
19 |
+
"6": "LABEL_6",
|
20 |
+
"7": "LABEL_7",
|
21 |
+
"8": "LABEL_8",
|
22 |
+
"9": "LABEL_9",
|
23 |
+
"10": "LABEL_10"
|
24 |
+
},
|
25 |
+
"initializer_range": 0.02,
|
26 |
+
"intermediate_size": 3072,
|
27 |
+
"label2id": {
|
28 |
+
"LABEL_0": 0,
|
29 |
+
"LABEL_1": 1,
|
30 |
+
"LABEL_10": 10,
|
31 |
+
"LABEL_2": 2,
|
32 |
+
"LABEL_3": 3,
|
33 |
+
"LABEL_4": 4,
|
34 |
+
"LABEL_5": 5,
|
35 |
+
"LABEL_6": 6,
|
36 |
+
"LABEL_7": 7,
|
37 |
+
"LABEL_8": 8,
|
38 |
+
"LABEL_9": 9
|
39 |
+
},
|
40 |
+
"layer_norm_eps": 1e-05,
|
41 |
+
"max_position_embeddings": 514,
|
42 |
+
"model_type": "roberta",
|
43 |
+
"num_attention_heads": 12,
|
44 |
+
"num_hidden_layers": 12,
|
45 |
+
"pad_token_id": 1,
|
46 |
+
"type_vocab_size": 1,
|
47 |
+
"vocab_size": 50265
|
48 |
+
}
|
eval_results.txt
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
LRAP = 0.9073448773448772
|
2 |
+
eval_loss = 0.17623897372099587
|
3 |
+
macro_f1 = 0.5891904448035862
|
4 |
+
micro_f1 = 0.6097087378640776
|
5 |
+
weighted_f1 = 0.60211948555892
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
model_args.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"adam_epsilon": 1e-08, "best_model_dir": "/shared/2/projects/framing/models/classify/Issue-Specific/11-03-20_60_epochs_default_thresh_12_seed/best_model", "cache_dir": "cache_dir/", "config": {}, "custom_layer_parameters": [], "custom_parameter_groups": [], "dataloader_num_workers": 14, "do_lower_case": false, "dynamic_quantize": false, "early_stopping_consider_epochs": false, "early_stopping_delta": 0, "early_stopping_metric": "eval_loss", "early_stopping_metric_minimize": true, "early_stopping_patience": 20, "encoding": null, "eval_batch_size": 8, "evaluate_during_training": true, "evaluate_during_training_silent": true, "evaluate_during_training_steps": 100, "evaluate_during_training_verbose": false, "fp16": false, "gradient_accumulation_steps": 1, "learning_rate": 4e-05, "local_rank": -1, "logging_steps": 50, "manual_seed": 12, "max_grad_norm": 1.0, "max_seq_length": 128, "model_name": "/shared/2/projects/framing/models/finetune/roberta_cased_09-01-20", "model_type": "roberta", "multiprocessing_chunksize": 500, "n_gpu": 1, "no_cache": false, "no_save": false, "num_train_epochs": 60, "output_dir": "/shared/2/projects/framing/models/classify/Issue-Specific/11-03-20_60_epochs_default_thresh_12_seed", "overwrite_output_dir": true, "process_count": 14, "quantized_model": false, "reprocess_input_data": true, "save_best_model": true, "save_eval_checkpoints": true, "save_model_every_epoch": true, "save_optimizer_and_scheduler": true, "save_steps": 2000, "silent": false, "tensorboard_dir": null, "thread_count": null, "train_batch_size": 8, "train_custom_parameters_only": false, "use_cached_eval_features": false, "use_early_stopping": true, "use_multiprocessing": true, "wandb_kwargs": {}, "wandb_project": null, "warmup_ratio": 0.06, "warmup_steps": 1620, "weight_decay": 0, "model_class": "MultiLabelClassificationModel", "sliding_window": false, "stride": 0.8, "threshold": 0.5, "tie_value": 1, "labels_list": [], "labels_map": {}, "lazy_loading": false}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:132de353ed16fe4322ab7ba53d4a6d771149c14e9788c4c4174c42ac5711d307
|
3 |
+
size 501070107
|
special_tokens_map.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>"}
|
tokenizer_config.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"model_max_length": 512, "do_lower_case": false, "special_tokens_map_file": "/shared/2/projects/framing/models/finetune/roberta_cased_09-01-20/special_tokens_map.json", "full_tokenizer_file": null}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:119431264a48ece94c94fbfbdb6c4df4db7b8a3d09eb45c68e81918ebd8ec82e
|
3 |
+
size 2479
|
training_progress_scores.csv
ADDED
@@ -0,0 +1,42 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
global_step,LRAP,train_loss,eval_loss,macro_f1,weighted_f1,micro_f1
|
2 |
+
100,0.7057642295975629,0.496387243270874,0.48415745768630714,0.006734006734006734,0.01182033096926714,0.013201320132013201
|
3 |
+
200,0.7532075917909248,0.2193269431591034,0.24408512680154099,0.0,0.0,0.0
|
4 |
+
300,0.7495410453743788,0.16921024024486542,0.2079820661691197,0.0,0.0,0.0
|
5 |
+
400,0.7646333974667301,0.19834978878498077,0.19780043891647406,0.0,0.0,0.0
|
6 |
+
450,0.7501008497675158,0.23687458038330078,0.19518689994226424,0.0,0.0,0.0
|
7 |
+
500,0.812143177809844,0.1969013214111328,0.18708720358840206,0.004784688995215312,0.013811123553564763,0.014084507042253521
|
8 |
+
600,0.854238255571589,0.2504305839538574,0.15873292990420995,0.06600410047939799,0.1806792921140237,0.2370820668693009
|
9 |
+
700,0.8664175084175078,0.13886785507202148,0.14376250314607955,0.1603330453659401,0.33752518732805226,0.3913043478260869
|
10 |
+
800,0.8833271604938269,0.18927375972270966,0.1345756352601344,0.2427620184061604,0.46649309548956835,0.5127020785219399
|
11 |
+
900,0.8760202821869485,0.08767407387495041,0.12995589191192075,0.2388654606376692,0.45438315520334444,0.505800464037123
|
12 |
+
900,0.8760202821869485,0.08767407387495041,0.12995589191192075,0.2388654606376692,0.45438315520334444,0.505800464037123
|
13 |
+
1000,0.8803555395222059,0.07628375291824341,0.12778916314505695,0.2936439544276605,0.4742139742964635,0.5141509433962264
|
14 |
+
1100,0.8972364117364113,0.07170897722244263,0.12786507668594518,0.3684848115854503,0.5079093252955212,0.5524193548387096
|
15 |
+
1200,0.8988625941959274,0.12749052047729492,0.12654653508542923,0.37010409966656516,0.518366473966867,0.5643153526970954
|
16 |
+
1300,0.8978597883597881,0.12220723181962967,0.12996221328840443,0.29102722228307987,0.48276679146935786,0.5371900826446282
|
17 |
+
1350,0.9080705467372132,0.06898022443056107,0.12048480564957126,0.42373125585299065,0.5514303916855643,0.5793991416309013
|
18 |
+
1400,0.8996622574955907,0.15102437138557434,0.1291346744748584,0.47012642765790846,0.5739814884945889,0.598848368522073
|
19 |
+
1500,0.9074727433060763,0.1877550333738327,0.13238610650755858,0.32227107357837415,0.4991849619473071,0.5333333333333333
|
20 |
+
1600,0.8925214045214044,0.010207578539848328,0.13011972047388554,0.4803555461839388,0.5782011443796453,0.5823293172690763
|
21 |
+
1700,0.897634038800705,0.012144484557211399,0.13760558408766724,0.3588330836380166,0.5146476946991526,0.5539714867617107
|
22 |
+
1800,0.9012152477152474,0.09005559235811234,0.14026125458379587,0.4509276706937007,0.5600896223244543,0.5817490494296578
|
23 |
+
1800,0.9012152477152474,0.09005559235811234,0.14026125458379587,0.4509276706937007,0.5600896223244543,0.5817490494296578
|
24 |
+
1900,0.8997700016033345,0.05511873587965965,0.13812682359364994,0.5282129489810252,0.5856380241333823,0.5897920604914935
|
25 |
+
2000,0.9065953984287314,0.08643753826618195,0.14112595059467772,0.5255089595713817,0.5939656994651844,0.6020761245674741
|
26 |
+
2100,0.8978112874779536,0.013076628558337688,0.15222296858892628,0.5231535586130122,0.5857019414188477,0.5988909426987061
|
27 |
+
2200,0.8963528138528136,0.051304515451192856,0.1447930874776814,0.4845279013950305,0.5752403666752719,0.591715976331361
|
28 |
+
2250,0.9036165624498954,0.03514162078499794,0.15457817688257547,0.49689307331808935,0.5784754579031515,0.5970149253731344
|
29 |
+
2300,0.8983405483405482,0.007065874058753252,0.1442033687132623,0.5244866497005399,0.5972208646905794,0.6078431372549019
|
30 |
+
2400,0.9018360590027255,0.03274355083703995,0.1543740569532179,0.5614868729914396,0.6109094740740802,0.6111111111111112
|
31 |
+
2500,0.9057963764630428,0.023506447672843933,0.15581955929825964,0.5408152825587984,0.5870922602657457,0.5884476534296028
|
32 |
+
2600,0.9007267115600444,0.026641204953193665,0.15825238153175042,0.5809368400767084,0.614178333945124,0.6180422264875239
|
33 |
+
2700,0.9169559884559879,0.002666956977918744,0.15324820701866165,0.5239152223916904,0.5517656902395385,0.5608695652173913
|
34 |
+
2700,0.9169559884559879,0.002666956977918744,0.15324820701866165,0.5239152223916904,0.5517656902395385,0.5608695652173913
|
35 |
+
2800,0.9028503286836619,0.08380335569381714,0.16174583353118546,0.5874766428469718,0.6095671025951968,0.6170212765957447
|
36 |
+
2900,0.9004524611191274,0.0050240918062627316,0.1604357653587501,0.5652634022562363,0.5995979231153512,0.6185185185185186
|
37 |
+
3000,0.907663887552776,0.06582244485616684,0.1632129770148088,0.5463875013498201,0.5731142234888612,0.5836575875486382
|
38 |
+
3100,0.9020264550264543,0.006758783478289843,0.16617869030635216,0.5345072766049483,0.5974799937272903,0.6101694915254238
|
39 |
+
3150,0.8976526374859707,0.011146184988319874,0.17138368233613538,0.5266310463166574,0.5738081362930063,0.5788423153692616
|
40 |
+
3200,0.9090308641975307,0.004416145384311676,0.1689870411432103,0.49770617587437277,0.5714415806588806,0.5924453280318092
|
41 |
+
3300,0.9080821709155042,0.004951728507876396,0.1641895531251896,0.5908781009047008,0.5944440383880821,0.605072463768116
|
42 |
+
3400,0.9073448773448772,0.01981344260275364,0.17623897372099587,0.5891904448035862,0.60211948555892,0.6097087378640776
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|