mahdiyar
commited on
Commit
•
e746f62
1
Parent(s):
cf17781
add model weights and configs
Browse files- added_tokens.json +1 -0
- cartography_config_dev.json +49 -0
- cartography_config_train.json +47 -0
- config.json +28 -0
- eval_metrics_mnli_dev_.json +1 -0
- eval_metrics_mnli_dev_in_training.json +9 -0
- eval_metrics_train.json +9 -0
- merges.txt +0 -0
- predictions_mnli_dev_.lst +0 -0
- predictions_mnli_dev_in_training.lst +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer_config.json +1 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{}
|
cartography_config_dev.json
ADDED
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"adam_epsilon": 1e-08,
|
3 |
+
"cache_dir": "",
|
4 |
+
"config_name": "",
|
5 |
+
"data_dir": "/content/pretrained/filtered/cartography_variability_0.33/MNLI",
|
6 |
+
"dev": null,
|
7 |
+
"device": "cuda",
|
8 |
+
"do_eval": true,
|
9 |
+
"do_lower_case": true,
|
10 |
+
"do_test": false,
|
11 |
+
"do_train": true,
|
12 |
+
"eval_all_checkpoints": false,
|
13 |
+
"eval_batch_size": 96,
|
14 |
+
"evaluate_during_training": true,
|
15 |
+
"evaluate_during_training_epoch": false,
|
16 |
+
"features_cache_dir": "/content/pretrained/filtered/cartography_variability_0.33/MNLI/cache_36891",
|
17 |
+
"fp16": false,
|
18 |
+
"fp16_opt_level": "01",
|
19 |
+
"gradient_accumulation_steps": 1,
|
20 |
+
"learning_rate": 3.66e-06,
|
21 |
+
"local_rank": -1,
|
22 |
+
"logging_steps": 1000,
|
23 |
+
"max_grad_norm": 1.0,
|
24 |
+
"max_seq_length": 128,
|
25 |
+
"max_steps": -1,
|
26 |
+
"model_name_or_path": "/content/pretrained/results/",
|
27 |
+
"model_type": "roberta",
|
28 |
+
"n_gpu": 1,
|
29 |
+
"no_cuda": false,
|
30 |
+
"num_train_epochs": 9,
|
31 |
+
"output_dir": "/content/results",
|
32 |
+
"output_mode": "classification",
|
33 |
+
"overwrite_cache": false,
|
34 |
+
"overwrite_output_dir": false,
|
35 |
+
"patience": 9,
|
36 |
+
"per_gpu_eval_batch_size": 96,
|
37 |
+
"per_gpu_train_batch_size": 64,
|
38 |
+
"save_steps": 0,
|
39 |
+
"seed": 36891,
|
40 |
+
"server_ip": "",
|
41 |
+
"server_port": "",
|
42 |
+
"task_name": "mnli",
|
43 |
+
"test": null,
|
44 |
+
"tokenizer_name": "",
|
45 |
+
"train": null,
|
46 |
+
"train_batch_size": 64,
|
47 |
+
"warmup_steps": 0,
|
48 |
+
"weight_decay": 0.0
|
49 |
+
}
|
cartography_config_train.json
ADDED
@@ -0,0 +1,47 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"adam_epsilon": 1e-08,
|
3 |
+
"cache_dir": "",
|
4 |
+
"config_name": "",
|
5 |
+
"data_dir": "/content/pretrained/filtered/cartography_variability_0.33/MNLI",
|
6 |
+
"dev": null,
|
7 |
+
"device": "cuda",
|
8 |
+
"do_eval": true,
|
9 |
+
"do_lower_case": true,
|
10 |
+
"do_test": false,
|
11 |
+
"do_train": true,
|
12 |
+
"eval_all_checkpoints": false,
|
13 |
+
"evaluate_during_training": true,
|
14 |
+
"evaluate_during_training_epoch": false,
|
15 |
+
"features_cache_dir": "/content/pretrained/filtered/cartography_variability_0.33/MNLI/cache_36891",
|
16 |
+
"fp16": false,
|
17 |
+
"fp16_opt_level": "01",
|
18 |
+
"gradient_accumulation_steps": 1,
|
19 |
+
"learning_rate": 3.66e-06,
|
20 |
+
"local_rank": -1,
|
21 |
+
"logging_steps": 1000,
|
22 |
+
"max_grad_norm": 1.0,
|
23 |
+
"max_seq_length": 128,
|
24 |
+
"max_steps": -1,
|
25 |
+
"model_name_or_path": "/content/pretrained/results/",
|
26 |
+
"model_type": "roberta",
|
27 |
+
"n_gpu": 1,
|
28 |
+
"no_cuda": false,
|
29 |
+
"num_train_epochs": 9,
|
30 |
+
"output_dir": "/content/results",
|
31 |
+
"output_mode": "classification",
|
32 |
+
"overwrite_cache": false,
|
33 |
+
"overwrite_output_dir": false,
|
34 |
+
"patience": 9,
|
35 |
+
"per_gpu_eval_batch_size": 96,
|
36 |
+
"per_gpu_train_batch_size": 64,
|
37 |
+
"save_steps": 0,
|
38 |
+
"seed": 36891,
|
39 |
+
"server_ip": "",
|
40 |
+
"server_port": "",
|
41 |
+
"task_name": "mnli",
|
42 |
+
"test": null,
|
43 |
+
"tokenizer_name": "",
|
44 |
+
"train": null,
|
45 |
+
"warmup_steps": 0,
|
46 |
+
"weight_decay": 0.0
|
47 |
+
}
|
config.json
ADDED
@@ -0,0 +1,28 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"architectures": [
|
3 |
+
"RobertaForMaskedLM"
|
4 |
+
],
|
5 |
+
"attention_probs_dropout_prob": 0.1,
|
6 |
+
"bos_token_id": 0,
|
7 |
+
"eos_token_id": 2,
|
8 |
+
"finetuning_task": "mnli",
|
9 |
+
"hidden_act": "gelu",
|
10 |
+
"hidden_dropout_prob": 0.1,
|
11 |
+
"hidden_size": 768,
|
12 |
+
"initializer_range": 0.02,
|
13 |
+
"intermediate_size": 3072,
|
14 |
+
"layer_norm_eps": 1e-05,
|
15 |
+
"max_position_embeddings": 514,
|
16 |
+
"model_type": "roberta",
|
17 |
+
"num_attention_heads": 12,
|
18 |
+
"num_hidden_layers": 12,
|
19 |
+
"num_labels": 3,
|
20 |
+
"output_attentions": false,
|
21 |
+
"output_hidden_states": false,
|
22 |
+
"pad_token_id": 1,
|
23 |
+
"pruned_heads": {},
|
24 |
+
"torchscript": false,
|
25 |
+
"type_vocab_size": 1,
|
26 |
+
"use_bfloat16": false,
|
27 |
+
"vocab_size": 50265
|
28 |
+
}
|
eval_metrics_mnli_dev_.json
ADDED
@@ -0,0 +1 @@
|
|
|
|
|
1 |
+
{"acc": 0.8760061130922058}
|
eval_metrics_mnli_dev_in_training.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{"acc": 0.8746816097809476}
|
2 |
+
{"acc": 0.8757004584819155}
|
3 |
+
{"acc": 0.8760061130922058}
|
4 |
+
{"acc": 0.8751910341314315}
|
5 |
+
{"acc": 0.8749872643912379}
|
6 |
+
{"acc": 0.8737646459500764}
|
7 |
+
{"acc": 0.8743759551706571}
|
8 |
+
{"acc": 0.8735608762098829}
|
9 |
+
{"acc": 0.87396841569027}
|
eval_metrics_train.json
ADDED
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{"epoch": 0, "train_acc": 0.9124507050898434, "best_dev_performance": 0.8746816097809476, "avg_batch_loss": 7.11664024181664, "learning_rate": 3.2533333333333332e-06}
|
2 |
+
{"epoch": 1, "train_acc": 0.9262839693719407, "best_dev_performance": 0.8757004584819155, "avg_batch_loss": 6.131773860484827, "learning_rate": 2.846666666666667e-06}
|
3 |
+
{"epoch": 2, "train_acc": 0.9340714143227609, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 5.4619441903196275, "learning_rate": 2.44e-06}
|
4 |
+
{"epoch": 3, "train_acc": 0.941505930733938, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 4.922465293901041, "learning_rate": 2.0333333333333335e-06}
|
5 |
+
{"epoch": 4, "train_acc": 0.9477282143350366, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 4.450159241823712, "learning_rate": 1.6266666666666666e-06}
|
6 |
+
{"epoch": 5, "train_acc": 0.9508585370344795, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 4.147213218006073, "learning_rate": 1.22e-06}
|
7 |
+
{"epoch": 6, "train_acc": 0.9543954947904678, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 3.8465021344600245, "learning_rate": 8.133333333333333e-07}
|
8 |
+
{"epoch": 7, "train_acc": 0.9573723702987617, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 3.6861731809622142, "learning_rate": 4.0666666666666666e-07}
|
9 |
+
{"epoch": 8, "train_acc": 0.9600193343460848, "best_dev_performance": 0.8760061130922058, "avg_batch_loss": 3.479075360490242, "learning_rate": 0.0}
|
merges.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|
predictions_mnli_dev_.lst
ADDED
The diff for this file is too large to render.
See raw diff
|
|
predictions_mnli_dev_in_training.lst
ADDED
The diff for this file is too large to render.
See raw diff
|
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:7f27aed8bb4d091c89bb3f421e9d00b213d261c009640392c972e6865d696996
|
3 |
+
size 501004048
|
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 |
+
{"do_lower_case": true, "max_len": 512, "bos_token": "<s>", "eos_token": "</s>", "unk_token": "<unk>", "sep_token": "</s>", "pad_token": "<pad>", "cls_token": "<s>", "mask_token": "<mask>", "init_inputs": []}
|
training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f734b2cd1e2b3f5535905586f3a844a97bf75a5e881f583777c68e9751ad4f9d
|
3 |
+
size 1487
|
vocab.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|