lsanochkin
commited on
Commit
•
df04b7a
1
Parent(s):
a5cb282
initial commit
Browse files- config.json +29 -0
- git_log.json +5 -0
- parameters.json +51 -0
- pytorch_model.bin +3 -0
- tokenizer.json +0 -0
- vocab.txt +0 -0
config.json
ADDED
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "serialization_dir/distilelectra.pth",
|
3 |
+
"architectures": [
|
4 |
+
"ElectraForPreTraining"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"classifier_dropout": null,
|
8 |
+
"embedding_size": 768,
|
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-12,
|
15 |
+
"max_position_embeddings": 512,
|
16 |
+
"model_type": "electra",
|
17 |
+
"num_attention_heads": 12,
|
18 |
+
"num_hidden_layers": 6,
|
19 |
+
"output_past": true,
|
20 |
+
"pad_token_id": 0,
|
21 |
+
"position_embedding_type": "absolute",
|
22 |
+
"summary_activation": "gelu",
|
23 |
+
"summary_last_dropout": 0.1,
|
24 |
+
"summary_type": "first",
|
25 |
+
"summary_use_proj": true,
|
26 |
+
"transformers_version": "4.11.3",
|
27 |
+
"type_vocab_size": 2,
|
28 |
+
"vocab_size": 30522
|
29 |
+
}
|
git_log.json
ADDED
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"repo_id": "<git.repo.base.Repo '/home/jovyan/distil_tmp/distillation/.git'>",
|
3 |
+
"repo_sha": "de6dcaeed8e16839757d0c5a3a5f7e60f0b454e6",
|
4 |
+
"repo_branch": "master"
|
5 |
+
}
|
parameters.json
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"force": true,
|
3 |
+
"dump_path": "./serialization_dir/distilbert_mse",
|
4 |
+
"data_file": "data/binarized_text.electra.pickle",
|
5 |
+
"student_type": "electra",
|
6 |
+
"student_config": "training_configs/distilelectra.json",
|
7 |
+
"student_pretrained_weights": "serialization_dir/distilelectra.pth",
|
8 |
+
"teacher_type": "electra",
|
9 |
+
"teacher_name": "google/electra-base-discriminator",
|
10 |
+
"temperature": 2.0,
|
11 |
+
"alpha_ce": 5.0,
|
12 |
+
"alpha_mlm": 2.0,
|
13 |
+
"alpha_clm": 0.0,
|
14 |
+
"alpha_mse": 1.0,
|
15 |
+
"alpha_cos": 1.0,
|
16 |
+
"mlm": true,
|
17 |
+
"mlm_mask_prop": 0.15,
|
18 |
+
"word_mask": 0.8,
|
19 |
+
"word_keep": 0.1,
|
20 |
+
"word_rand": 0.1,
|
21 |
+
"mlm_smoothing": 0.7,
|
22 |
+
"token_counts": "data/token_counts.electra.pickle",
|
23 |
+
"restrict_ce_to_mask": false,
|
24 |
+
"freeze_pos_embs": true,
|
25 |
+
"freeze_token_type_embds": false,
|
26 |
+
"n_epoch": 50,
|
27 |
+
"batch_size": 5,
|
28 |
+
"group_by_size": true,
|
29 |
+
"gradient_accumulation_steps": 50,
|
30 |
+
"warmup_prop": 0.05,
|
31 |
+
"weight_decay": 0.0,
|
32 |
+
"learning_rate": 0.0005,
|
33 |
+
"adam_epsilon": 1e-06,
|
34 |
+
"max_grad_norm": 5.0,
|
35 |
+
"initializer_range": 0.02,
|
36 |
+
"fp16": false,
|
37 |
+
"fp16_opt_level": "O1",
|
38 |
+
"n_gpu": 4,
|
39 |
+
"local_rank": 0,
|
40 |
+
"seed": 56,
|
41 |
+
"log_interval": 500,
|
42 |
+
"checkpoint_interval": 4000,
|
43 |
+
"world_size": 4,
|
44 |
+
"n_gpu_per_node": 4,
|
45 |
+
"global_rank": 0,
|
46 |
+
"n_nodes": 1,
|
47 |
+
"node_id": 0,
|
48 |
+
"multi_gpu": true,
|
49 |
+
"is_master": true,
|
50 |
+
"multi_node": false
|
51 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1d83dac7e383e19c6686bcf7294ad3477ce12176e95c697009fb7c2d0564ce0c
|
3 |
+
size 267998066
|
tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
vocab.txt
ADDED
The diff for this file is too large to render.
See raw diff
|
|