Update text encoder model
Browse files- config.json +1 -1
- output_linear.bin +1 -1
- pytorch_model.bin +1 -1
- tokenizer_config.json +1 -1
- training_args.json → training_args_1.json +0 -0
- training_args_2.json +20 -0
- training_args_3.json +20 -0
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "sonoisa/clip-
|
3 |
"architectures": [
|
4 |
"BertModel"
|
5 |
],
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "sonoisa/clip-vit-b-32-japanese",
|
3 |
"architectures": [
|
4 |
"BertModel"
|
5 |
],
|
output_linear.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 4721639
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:4da2aadf0a1303ba85d0834450c14c8a396cd43fdb11fb182dcdfd47989a2986
|
3 |
size 4721639
|
pytorch_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 442547953
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:1f78cd6a86b5a818d2971418d3f073417e38c2296aaea36110dcf4e474f8faa9
|
3 |
size 442547953
|
tokenizer_config.json
CHANGED
@@ -1 +1 @@
|
|
1 |
-
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "do_lower_case": true, "do_word_tokenize": true, "do_subword_tokenize": true, "word_tokenizer_type": "mecab", "subword_tokenizer_type": "wordpiece", "never_split": null, "mecab_kwargs": null, "is_fast": true, "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "sonoisa/clip-
|
|
|
1 |
+
{"unk_token": "[UNK]", "sep_token": "[SEP]", "pad_token": "[PAD]", "cls_token": "[CLS]", "mask_token": "[MASK]", "do_lower_case": true, "do_word_tokenize": true, "do_subword_tokenize": true, "word_tokenizer_type": "mecab", "subword_tokenizer_type": "wordpiece", "never_split": null, "mecab_kwargs": null, "is_fast": true, "model_max_length": 512, "special_tokens_map_file": null, "tokenizer_file": null, "name_or_path": "sonoisa/clip-vit-b-32-japanese", "tokenizer_class": "BertJapaneseTokenizer"}
|
training_args.json → training_args_1.json
RENAMED
File without changes
|
training_args_2.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"adam_epsilon": 1e-08,
|
3 |
+
"data_dir": "/content/data",
|
4 |
+
"early_stop_callback": false,
|
5 |
+
"eval_batch_size": 32,
|
6 |
+
"fp_16": false,
|
7 |
+
"gradient_accumulation_steps": 4,
|
8 |
+
"learning_rate": 0.0002,
|
9 |
+
"max_grad_norm": 1.0,
|
10 |
+
"max_input_length": 64,
|
11 |
+
"model_name_or_path": "/content/pretrain_model",
|
12 |
+
"n_gpu": 1,
|
13 |
+
"num_train_epochs": 2,
|
14 |
+
"seed": 42,
|
15 |
+
"shuffle_buffer_size": 65536,
|
16 |
+
"tokenizer_name_or_path": "/content/pretrain_model",
|
17 |
+
"train_batch_size": 256,
|
18 |
+
"warmup_ratio": 0.05,
|
19 |
+
"weight_decay": 0.0
|
20 |
+
}
|
training_args_3.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"adam_epsilon": 1e-08,
|
3 |
+
"data_dir": "/content/data",
|
4 |
+
"early_stop_callback": false,
|
5 |
+
"eval_batch_size": 32,
|
6 |
+
"fp_16": false,
|
7 |
+
"gradient_accumulation_steps": 4,
|
8 |
+
"learning_rate": 0.0002,
|
9 |
+
"max_grad_norm": 1.0,
|
10 |
+
"max_input_length": 64,
|
11 |
+
"model_name_or_path": "/content/pretrain_model",
|
12 |
+
"n_gpu": 1,
|
13 |
+
"num_train_epochs": 2,
|
14 |
+
"seed": 12345,
|
15 |
+
"shuffle_buffer_size": 65536,
|
16 |
+
"tokenizer_name_or_path": "/content/pretrain_model",
|
17 |
+
"train_batch_size": 256,
|
18 |
+
"warmup_ratio": 0.05,
|
19 |
+
"weight_decay": 0.0
|
20 |
+
}
|