labicquette
commited on
Commit
•
e3b2bba
1
Parent(s):
ef0016d
upload models
Browse files- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/added_tokens.json +3 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/config.json +39 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/model.safetensors +3 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/sentencepiece.bpe.model +3 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/special_tokens_map.json +20 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/tokenizer.json +0 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/tokenizer_config.json +85 -0
- almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/training_args.bin +3 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/config.json +39 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/model.safetensors +3 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/optimizer.pt +3 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/rng_state.pth +3 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/scheduler.pt +3 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/trainer_state.json +108 -0
- almanach_camembert-base-ft-BUILD/checkpoint-1872/training_args.bin +3 -0
- almanach_camembert-base-ft-BUILD/logs/events.out.tfevents.1710368171.DESKTOP-VOM7ARG.126446.0 +3 -0
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/added_tokens.json
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"<unk>NOTUSED": 32005
|
3 |
+
}
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/config.json
ADDED
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "almanach/camembert-base",
|
3 |
+
"architectures": [
|
4 |
+
"CamembertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 5,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_id": 6,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"id2label": {
|
14 |
+
"0": "Entr\u00e9e",
|
15 |
+
"1": "Plat principal",
|
16 |
+
"2": "Dessert"
|
17 |
+
},
|
18 |
+
"initializer_range": 0.02,
|
19 |
+
"intermediate_size": 3072,
|
20 |
+
"label2id": {
|
21 |
+
"Dessert": 2,
|
22 |
+
"Entr\u00e9e": 0,
|
23 |
+
"Plat principal": 1
|
24 |
+
},
|
25 |
+
"layer_norm_eps": 1e-05,
|
26 |
+
"max_position_embeddings": 514,
|
27 |
+
"model_type": "camembert",
|
28 |
+
"num_attention_heads": 12,
|
29 |
+
"num_hidden_layers": 12,
|
30 |
+
"output_past": true,
|
31 |
+
"pad_token_id": 1,
|
32 |
+
"position_embedding_type": "absolute",
|
33 |
+
"problem_type": "single_label_classification",
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.36.1",
|
36 |
+
"type_vocab_size": 1,
|
37 |
+
"use_cache": true,
|
38 |
+
"vocab_size": 32005
|
39 |
+
}
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5fd1a1e03cc42543683f360d39c007ae7155483dc264f1b1149a21486364f41e
|
3 |
+
size 442521180
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/sentencepiece.bpe.model
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:988bc5a00281c6d210a5d34bd143d0363741a432fefe741bf71e61b1869d4314
|
3 |
+
size 810912
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/special_tokens_map.json
ADDED
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"additional_special_tokens": [
|
3 |
+
"<s>NOTUSED",
|
4 |
+
"</s>NOTUSED",
|
5 |
+
"<unk>NOTUSED"
|
6 |
+
],
|
7 |
+
"bos_token": "<s>",
|
8 |
+
"cls_token": "<s>",
|
9 |
+
"eos_token": "</s>",
|
10 |
+
"mask_token": {
|
11 |
+
"content": "<mask>",
|
12 |
+
"lstrip": true,
|
13 |
+
"normalized": false,
|
14 |
+
"rstrip": false,
|
15 |
+
"single_word": false
|
16 |
+
},
|
17 |
+
"pad_token": "<pad>",
|
18 |
+
"sep_token": "</s>",
|
19 |
+
"unk_token": "<unk>"
|
20 |
+
}
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/tokenizer.json
ADDED
The diff for this file is too large to render.
See raw diff
|
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/tokenizer_config.json
ADDED
@@ -0,0 +1,85 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"added_tokens_decoder": {
|
3 |
+
"0": {
|
4 |
+
"content": "<s>NOTUSED",
|
5 |
+
"lstrip": false,
|
6 |
+
"normalized": false,
|
7 |
+
"rstrip": false,
|
8 |
+
"single_word": false,
|
9 |
+
"special": true
|
10 |
+
},
|
11 |
+
"1": {
|
12 |
+
"content": "<pad>",
|
13 |
+
"lstrip": false,
|
14 |
+
"normalized": false,
|
15 |
+
"rstrip": false,
|
16 |
+
"single_word": false,
|
17 |
+
"special": true
|
18 |
+
},
|
19 |
+
"2": {
|
20 |
+
"content": "</s>NOTUSED",
|
21 |
+
"lstrip": false,
|
22 |
+
"normalized": false,
|
23 |
+
"rstrip": false,
|
24 |
+
"single_word": false,
|
25 |
+
"special": true
|
26 |
+
},
|
27 |
+
"4": {
|
28 |
+
"content": "<unk>",
|
29 |
+
"lstrip": false,
|
30 |
+
"normalized": false,
|
31 |
+
"rstrip": false,
|
32 |
+
"single_word": false,
|
33 |
+
"special": true
|
34 |
+
},
|
35 |
+
"5": {
|
36 |
+
"content": "<s>",
|
37 |
+
"lstrip": false,
|
38 |
+
"normalized": false,
|
39 |
+
"rstrip": false,
|
40 |
+
"single_word": false,
|
41 |
+
"special": true
|
42 |
+
},
|
43 |
+
"6": {
|
44 |
+
"content": "</s>",
|
45 |
+
"lstrip": false,
|
46 |
+
"normalized": false,
|
47 |
+
"rstrip": false,
|
48 |
+
"single_word": false,
|
49 |
+
"special": true
|
50 |
+
},
|
51 |
+
"32004": {
|
52 |
+
"content": "<mask>",
|
53 |
+
"lstrip": true,
|
54 |
+
"normalized": false,
|
55 |
+
"rstrip": false,
|
56 |
+
"single_word": false,
|
57 |
+
"special": true
|
58 |
+
},
|
59 |
+
"32005": {
|
60 |
+
"content": "<unk>NOTUSED",
|
61 |
+
"lstrip": false,
|
62 |
+
"normalized": false,
|
63 |
+
"rstrip": false,
|
64 |
+
"single_word": false,
|
65 |
+
"special": true
|
66 |
+
}
|
67 |
+
},
|
68 |
+
"additional_special_tokens": [
|
69 |
+
"<s>NOTUSED",
|
70 |
+
"</s>NOTUSED",
|
71 |
+
"<unk>NOTUSED"
|
72 |
+
],
|
73 |
+
"bos_token": "<s>",
|
74 |
+
"clean_up_tokenization_spaces": true,
|
75 |
+
"cls_token": "<s>",
|
76 |
+
"eos_token": "</s>",
|
77 |
+
"mask_token": "<mask>",
|
78 |
+
"model_max_length": 512,
|
79 |
+
"pad_token": "<pad>",
|
80 |
+
"padding": "max_length",
|
81 |
+
"sep_token": "</s>",
|
82 |
+
"tokenizer_class": "CamembertTokenizer",
|
83 |
+
"truncation": true,
|
84 |
+
"unk_token": "<unk>"
|
85 |
+
}
|
almanach_camembert-base-ft-BUILD/almanach_camembert-base-ft-BUILD-best/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5c31870ced0a8f4d94fd51e68b1a3d99cfa79aae3948a0590a8a8383a5f76d2a
|
3 |
+
size 4856
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/config.json
ADDED
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"_name_or_path": "almanach/camembert-base",
|
3 |
+
"architectures": [
|
4 |
+
"CamembertForSequenceClassification"
|
5 |
+
],
|
6 |
+
"attention_probs_dropout_prob": 0.1,
|
7 |
+
"bos_token_id": 5,
|
8 |
+
"classifier_dropout": null,
|
9 |
+
"eos_token_id": 6,
|
10 |
+
"hidden_act": "gelu",
|
11 |
+
"hidden_dropout_prob": 0.1,
|
12 |
+
"hidden_size": 768,
|
13 |
+
"id2label": {
|
14 |
+
"0": "Entr\u00e9e",
|
15 |
+
"1": "Plat principal",
|
16 |
+
"2": "Dessert"
|
17 |
+
},
|
18 |
+
"initializer_range": 0.02,
|
19 |
+
"intermediate_size": 3072,
|
20 |
+
"label2id": {
|
21 |
+
"Dessert": 2,
|
22 |
+
"Entr\u00e9e": 0,
|
23 |
+
"Plat principal": 1
|
24 |
+
},
|
25 |
+
"layer_norm_eps": 1e-05,
|
26 |
+
"max_position_embeddings": 514,
|
27 |
+
"model_type": "camembert",
|
28 |
+
"num_attention_heads": 12,
|
29 |
+
"num_hidden_layers": 12,
|
30 |
+
"output_past": true,
|
31 |
+
"pad_token_id": 1,
|
32 |
+
"position_embedding_type": "absolute",
|
33 |
+
"problem_type": "single_label_classification",
|
34 |
+
"torch_dtype": "float32",
|
35 |
+
"transformers_version": "4.36.1",
|
36 |
+
"type_vocab_size": 1,
|
37 |
+
"use_cache": true,
|
38 |
+
"vocab_size": 32005
|
39 |
+
}
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/model.safetensors
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:df3fec3c36512600cdc4bc862200cacb549c17a08438d43b46d65383cddf7f49
|
3 |
+
size 442521180
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/optimizer.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:2d377c7e893f5c33df0f627756987dc662575e1ddb85506ebc451bbde10dd44d
|
3 |
+
size 885162234
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/rng_state.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a8cea68682beddf6bbb5ddcb716088687046bb2ba50492ed613ecb91c6eebac0
|
3 |
+
size 14244
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/scheduler.pt
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c1440c9e10d13ff14466df76139d08d1ce67c8903a7e53e0afbe6dcf6c40fc97
|
3 |
+
size 1064
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/trainer_state.json
ADDED
@@ -0,0 +1,108 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"best_metric": 0.886397407064587,
|
3 |
+
"best_model_checkpoint": "./models/camembert-base-combination/almanach_camembert-base-ft-BUILD/checkpoint-1248",
|
4 |
+
"epoch": 3.0,
|
5 |
+
"eval_steps": 500,
|
6 |
+
"global_step": 1872,
|
7 |
+
"is_hyper_param_search": false,
|
8 |
+
"is_local_process_zero": true,
|
9 |
+
"is_world_process_zero": true,
|
10 |
+
"log_history": [
|
11 |
+
{
|
12 |
+
"epoch": 0.32,
|
13 |
+
"learning_rate": 2e-05,
|
14 |
+
"loss": 0.824,
|
15 |
+
"step": 200
|
16 |
+
},
|
17 |
+
{
|
18 |
+
"epoch": 0.64,
|
19 |
+
"learning_rate": 4e-05,
|
20 |
+
"loss": 0.3968,
|
21 |
+
"step": 400
|
22 |
+
},
|
23 |
+
{
|
24 |
+
"epoch": 0.96,
|
25 |
+
"learning_rate": 4.845869297163995e-05,
|
26 |
+
"loss": 0.3408,
|
27 |
+
"step": 600
|
28 |
+
},
|
29 |
+
{
|
30 |
+
"epoch": 1.0,
|
31 |
+
"eval_f1": 0.8173488973093911,
|
32 |
+
"eval_loss": 0.49443894624710083,
|
33 |
+
"eval_p": 0.8435393842560946,
|
34 |
+
"eval_r": 0.8080929487179487,
|
35 |
+
"eval_runtime": 129.7007,
|
36 |
+
"eval_samples_per_second": 19.244,
|
37 |
+
"eval_steps_per_second": 1.203,
|
38 |
+
"step": 624
|
39 |
+
},
|
40 |
+
{
|
41 |
+
"epoch": 1.28,
|
42 |
+
"learning_rate": 4.5376078914919854e-05,
|
43 |
+
"loss": 0.2968,
|
44 |
+
"step": 800
|
45 |
+
},
|
46 |
+
{
|
47 |
+
"epoch": 1.6,
|
48 |
+
"learning_rate": 4.2293464858199754e-05,
|
49 |
+
"loss": 0.2811,
|
50 |
+
"step": 1000
|
51 |
+
},
|
52 |
+
{
|
53 |
+
"epoch": 1.92,
|
54 |
+
"learning_rate": 3.921085080147966e-05,
|
55 |
+
"loss": 0.3018,
|
56 |
+
"step": 1200
|
57 |
+
},
|
58 |
+
{
|
59 |
+
"epoch": 2.0,
|
60 |
+
"eval_f1": 0.886397407064587,
|
61 |
+
"eval_loss": 0.30126821994781494,
|
62 |
+
"eval_p": 0.8860404348357003,
|
63 |
+
"eval_r": 0.8870192307692307,
|
64 |
+
"eval_runtime": 141.7944,
|
65 |
+
"eval_samples_per_second": 17.603,
|
66 |
+
"eval_steps_per_second": 1.1,
|
67 |
+
"step": 1248
|
68 |
+
},
|
69 |
+
{
|
70 |
+
"epoch": 2.24,
|
71 |
+
"learning_rate": 3.6128236744759556e-05,
|
72 |
+
"loss": 0.254,
|
73 |
+
"step": 1400
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"epoch": 2.56,
|
77 |
+
"learning_rate": 3.304562268803946e-05,
|
78 |
+
"loss": 0.228,
|
79 |
+
"step": 1600
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"epoch": 2.88,
|
83 |
+
"learning_rate": 2.996300863131936e-05,
|
84 |
+
"loss": 0.2307,
|
85 |
+
"step": 1800
|
86 |
+
},
|
87 |
+
{
|
88 |
+
"epoch": 3.0,
|
89 |
+
"eval_f1": 0.8839271082580393,
|
90 |
+
"eval_loss": 0.302884578704834,
|
91 |
+
"eval_p": 0.8846466611130432,
|
92 |
+
"eval_r": 0.8854166666666666,
|
93 |
+
"eval_runtime": 139.7336,
|
94 |
+
"eval_samples_per_second": 17.863,
|
95 |
+
"eval_steps_per_second": 1.116,
|
96 |
+
"step": 1872
|
97 |
+
}
|
98 |
+
],
|
99 |
+
"logging_steps": 200,
|
100 |
+
"max_steps": 3744,
|
101 |
+
"num_input_tokens_seen": 0,
|
102 |
+
"num_train_epochs": 6,
|
103 |
+
"save_steps": 500,
|
104 |
+
"total_flos": 7875247706016768.0,
|
105 |
+
"train_batch_size": 16,
|
106 |
+
"trial_name": null,
|
107 |
+
"trial_params": null
|
108 |
+
}
|
almanach_camembert-base-ft-BUILD/checkpoint-1872/training_args.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5c31870ced0a8f4d94fd51e68b1a3d99cfa79aae3948a0590a8a8383a5f76d2a
|
3 |
+
size 4856
|
almanach_camembert-base-ft-BUILD/logs/events.out.tfevents.1710368171.DESKTOP-VOM7ARG.126446.0
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b83d7c6f59dd09f6aaf5b792c64a9853aeec9572e09970aa03bbe3c70b5ef55c
|
3 |
+
size 10364
|