Upload folder using huggingface_hub
Browse files- checkpoint-3126/config.json +54 -0
- checkpoint-3126/merges.txt +0 -0
- checkpoint-3126/model.safetensors +3 -0
- checkpoint-3126/optimizer.pt +3 -0
- checkpoint-3126/rng_state.pth +3 -0
- checkpoint-3126/scheduler.pt +3 -0
- checkpoint-3126/special_tokens_map.json +51 -0
- checkpoint-3126/tokenizer_config.json +57 -0
- checkpoint-3126/trainer_state.json +286 -0
- checkpoint-3126/training_args.bin +3 -0
- checkpoint-3126/vocab.json +0 -0
- checkpoint-4689/config.json +54 -0
- checkpoint-4689/merges.txt +0 -0
- checkpoint-4689/model.safetensors +3 -0
- checkpoint-4689/optimizer.pt +3 -0
- checkpoint-4689/rng_state.pth +3 -0
- checkpoint-4689/scheduler.pt +3 -0
- checkpoint-4689/special_tokens_map.json +51 -0
- checkpoint-4689/tokenizer_config.json +57 -0
- checkpoint-4689/trainer_state.json +404 -0
- checkpoint-4689/training_args.bin +3 -0
- checkpoint-4689/vocab.json +0 -0
- config.json +54 -0
- eval_metrics.json +9 -0
- merges.txt +0 -0
- model.safetensors +3 -0
- special_tokens_map.json +51 -0
- tokenizer_config.json +57 -0
- training_args.bin +3 -0
- vocab.json +0 -0
checkpoint-3126/config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 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 |
+
"output_past": true,
|
| 46 |
+
"pad_token_id": 1,
|
| 47 |
+
"position_embedding_type": "absolute",
|
| 48 |
+
"problem_type": "single_label_classification",
|
| 49 |
+
"torch_dtype": "float32",
|
| 50 |
+
"transformers_version": "4.53.3",
|
| 51 |
+
"type_vocab_size": 1,
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"vocab_size": 50265
|
| 54 |
+
}
|
checkpoint-3126/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-3126/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4ac3c3a499ce0b5015aba3865a1b16fadfab8d4689c48beaa72c6624bfeed71b
|
| 3 |
+
size 498640508
|
checkpoint-3126/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:56ac9b1dd751f8145c29ccad0bfd3e32f7e238d4e55c371ae6e9c23123a52bb7
|
| 3 |
+
size 997400762
|
checkpoint-3126/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:877c2662e7454e081572b0c9955ccc79db48d77d85879e0865a5f2c23ccd35b3
|
| 3 |
+
size 14244
|
checkpoint-3126/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2db203eb681715c2ed924110e180ee12d2eafac6b5c46b7e8c775d7e8d92d35c
|
| 3 |
+
size 1064
|
checkpoint-3126/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-3126/tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
checkpoint-3126/trainer_state.json
ADDED
|
@@ -0,0 +1,286 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 3126,
|
| 3 |
+
"best_metric": 0.284542217831853,
|
| 4 |
+
"best_model_checkpoint": "/kaggle/working/codebert_b_local/checkpoint-3126",
|
| 5 |
+
"epoch": 2.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 3126,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.06397952655150352,
|
| 14 |
+
"grad_norm": 4.3835320472717285,
|
| 15 |
+
"learning_rate": 3.96e-06,
|
| 16 |
+
"loss": 1.6251,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.12795905310300704,
|
| 21 |
+
"grad_norm": 1.2909411191940308,
|
| 22 |
+
"learning_rate": 7.960000000000002e-06,
|
| 23 |
+
"loss": 0.6209,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.19193857965451055,
|
| 28 |
+
"grad_norm": 7.392430305480957,
|
| 29 |
+
"learning_rate": 1.196e-05,
|
| 30 |
+
"loss": 0.5322,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.2559181062060141,
|
| 35 |
+
"grad_norm": 2.060398578643799,
|
| 36 |
+
"learning_rate": 1.5960000000000003e-05,
|
| 37 |
+
"loss": 0.4804,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.3198976327575176,
|
| 42 |
+
"grad_norm": 19.443620681762695,
|
| 43 |
+
"learning_rate": 1.9960000000000002e-05,
|
| 44 |
+
"loss": 0.4048,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.3838771593090211,
|
| 49 |
+
"grad_norm": 12.475482940673828,
|
| 50 |
+
"learning_rate": 1.9527333492480306e-05,
|
| 51 |
+
"loss": 0.4219,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.44785668586052463,
|
| 56 |
+
"grad_norm": 7.762031555175781,
|
| 57 |
+
"learning_rate": 1.9049892575793746e-05,
|
| 58 |
+
"loss": 0.3789,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.5118362124120281,
|
| 63 |
+
"grad_norm": 2.098180055618286,
|
| 64 |
+
"learning_rate": 1.8572451659107187e-05,
|
| 65 |
+
"loss": 0.4134,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.5758157389635317,
|
| 70 |
+
"grad_norm": 2.177507162094116,
|
| 71 |
+
"learning_rate": 1.8095010742420627e-05,
|
| 72 |
+
"loss": 0.3903,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.6397952655150352,
|
| 77 |
+
"grad_norm": 5.113386631011963,
|
| 78 |
+
"learning_rate": 1.7617569825734067e-05,
|
| 79 |
+
"loss": 0.3559,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.7037747920665387,
|
| 84 |
+
"grad_norm": 4.909190654754639,
|
| 85 |
+
"learning_rate": 1.7140128909047508e-05,
|
| 86 |
+
"loss": 0.3219,
|
| 87 |
+
"step": 1100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.7677543186180422,
|
| 91 |
+
"grad_norm": 5.605203628540039,
|
| 92 |
+
"learning_rate": 1.6662687992360948e-05,
|
| 93 |
+
"loss": 0.3821,
|
| 94 |
+
"step": 1200
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.8317338451695457,
|
| 98 |
+
"grad_norm": 6.7283453941345215,
|
| 99 |
+
"learning_rate": 1.6185247075674385e-05,
|
| 100 |
+
"loss": 0.3674,
|
| 101 |
+
"step": 1300
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.8957133717210493,
|
| 105 |
+
"grad_norm": 4.659298419952393,
|
| 106 |
+
"learning_rate": 1.5707806158987825e-05,
|
| 107 |
+
"loss": 0.3485,
|
| 108 |
+
"step": 1400
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.9596928982725528,
|
| 112 |
+
"grad_norm": 6.4099931716918945,
|
| 113 |
+
"learning_rate": 1.5230365242301266e-05,
|
| 114 |
+
"loss": 0.3565,
|
| 115 |
+
"step": 1500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.0,
|
| 119 |
+
"eval_accuracy": 0.9024,
|
| 120 |
+
"eval_f1_macro": 0.23154395234377326,
|
| 121 |
+
"eval_f1_weighted": 0.8916810274029613,
|
| 122 |
+
"eval_loss": 0.3302375078201294,
|
| 123 |
+
"eval_precision_macro": 0.2442325389892706,
|
| 124 |
+
"eval_recall_macro": 0.2618191076765516,
|
| 125 |
+
"eval_runtime": 91.0613,
|
| 126 |
+
"eval_samples_per_second": 109.816,
|
| 127 |
+
"eval_steps_per_second": 3.437,
|
| 128 |
+
"step": 1563
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.0236724248240563,
|
| 132 |
+
"grad_norm": 5.717738151550293,
|
| 133 |
+
"learning_rate": 1.4752924325614706e-05,
|
| 134 |
+
"loss": 0.318,
|
| 135 |
+
"step": 1600
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.0876519513755598,
|
| 139 |
+
"grad_norm": 4.969061851501465,
|
| 140 |
+
"learning_rate": 1.4275483408928146e-05,
|
| 141 |
+
"loss": 0.3001,
|
| 142 |
+
"step": 1700
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.1516314779270633,
|
| 146 |
+
"grad_norm": 3.3130006790161133,
|
| 147 |
+
"learning_rate": 1.3798042492241585e-05,
|
| 148 |
+
"loss": 0.3332,
|
| 149 |
+
"step": 1800
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.2156110044785668,
|
| 153 |
+
"grad_norm": 4.077260971069336,
|
| 154 |
+
"learning_rate": 1.3320601575555025e-05,
|
| 155 |
+
"loss": 0.32,
|
| 156 |
+
"step": 1900
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.2795905310300704,
|
| 160 |
+
"grad_norm": 3.1868574619293213,
|
| 161 |
+
"learning_rate": 1.2843160658868466e-05,
|
| 162 |
+
"loss": 0.3095,
|
| 163 |
+
"step": 2000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.3435700575815739,
|
| 167 |
+
"grad_norm": 6.247142314910889,
|
| 168 |
+
"learning_rate": 1.2365719742181906e-05,
|
| 169 |
+
"loss": 0.3105,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.4075495841330774,
|
| 174 |
+
"grad_norm": 11.029816627502441,
|
| 175 |
+
"learning_rate": 1.1888278825495345e-05,
|
| 176 |
+
"loss": 0.2975,
|
| 177 |
+
"step": 2200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.471529110684581,
|
| 181 |
+
"grad_norm": 1.82353937625885,
|
| 182 |
+
"learning_rate": 1.1410837908808785e-05,
|
| 183 |
+
"loss": 0.3047,
|
| 184 |
+
"step": 2300
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.5355086372360844,
|
| 188 |
+
"grad_norm": 8.549129486083984,
|
| 189 |
+
"learning_rate": 1.0933396992122225e-05,
|
| 190 |
+
"loss": 0.2983,
|
| 191 |
+
"step": 2400
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.599488163787588,
|
| 195 |
+
"grad_norm": 5.218935966491699,
|
| 196 |
+
"learning_rate": 1.0455956075435664e-05,
|
| 197 |
+
"loss": 0.2965,
|
| 198 |
+
"step": 2500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.6634676903390915,
|
| 202 |
+
"grad_norm": 1.3732364177703857,
|
| 203 |
+
"learning_rate": 9.978515158749106e-06,
|
| 204 |
+
"loss": 0.274,
|
| 205 |
+
"step": 2600
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.727447216890595,
|
| 209 |
+
"grad_norm": 5.042181968688965,
|
| 210 |
+
"learning_rate": 9.501074242062547e-06,
|
| 211 |
+
"loss": 0.2804,
|
| 212 |
+
"step": 2700
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.7914267434420985,
|
| 216 |
+
"grad_norm": 2.0916764736175537,
|
| 217 |
+
"learning_rate": 9.023633325375985e-06,
|
| 218 |
+
"loss": 0.2689,
|
| 219 |
+
"step": 2800
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.855406269993602,
|
| 223 |
+
"grad_norm": 5.730693817138672,
|
| 224 |
+
"learning_rate": 8.546192408689426e-06,
|
| 225 |
+
"loss": 0.2798,
|
| 226 |
+
"step": 2900
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.9193857965451055,
|
| 230 |
+
"grad_norm": 1.510636806488037,
|
| 231 |
+
"learning_rate": 8.068751492002866e-06,
|
| 232 |
+
"loss": 0.2811,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.983365323096609,
|
| 237 |
+
"grad_norm": 5.20271110534668,
|
| 238 |
+
"learning_rate": 7.591310575316305e-06,
|
| 239 |
+
"loss": 0.2862,
|
| 240 |
+
"step": 3100
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.0,
|
| 244 |
+
"eval_accuracy": 0.9119,
|
| 245 |
+
"eval_f1_macro": 0.284542217831853,
|
| 246 |
+
"eval_f1_weighted": 0.900679083795208,
|
| 247 |
+
"eval_loss": 0.30098485946655273,
|
| 248 |
+
"eval_precision_macro": 0.36539760346541894,
|
| 249 |
+
"eval_recall_macro": 0.2947437498965655,
|
| 250 |
+
"eval_runtime": 91.3907,
|
| 251 |
+
"eval_samples_per_second": 109.42,
|
| 252 |
+
"eval_steps_per_second": 3.425,
|
| 253 |
+
"step": 3126
|
| 254 |
+
}
|
| 255 |
+
],
|
| 256 |
+
"logging_steps": 100,
|
| 257 |
+
"max_steps": 4689,
|
| 258 |
+
"num_input_tokens_seen": 0,
|
| 259 |
+
"num_train_epochs": 3,
|
| 260 |
+
"save_steps": 500,
|
| 261 |
+
"stateful_callbacks": {
|
| 262 |
+
"EarlyStoppingCallback": {
|
| 263 |
+
"args": {
|
| 264 |
+
"early_stopping_patience": 3,
|
| 265 |
+
"early_stopping_threshold": 0.0
|
| 266 |
+
},
|
| 267 |
+
"attributes": {
|
| 268 |
+
"early_stopping_patience_counter": 0
|
| 269 |
+
}
|
| 270 |
+
},
|
| 271 |
+
"TrainerControl": {
|
| 272 |
+
"args": {
|
| 273 |
+
"should_epoch_stop": false,
|
| 274 |
+
"should_evaluate": false,
|
| 275 |
+
"should_log": false,
|
| 276 |
+
"should_save": true,
|
| 277 |
+
"should_training_stop": false
|
| 278 |
+
},
|
| 279 |
+
"attributes": {}
|
| 280 |
+
}
|
| 281 |
+
},
|
| 282 |
+
"total_flos": 1.31566158336e+16,
|
| 283 |
+
"train_batch_size": 32,
|
| 284 |
+
"trial_name": null,
|
| 285 |
+
"trial_params": null
|
| 286 |
+
}
|
checkpoint-3126/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c79d0f0062a44424d0920ddb0ddfbfcf6ff0ddf33ebfac6155bad8cb9c616ba9
|
| 3 |
+
size 5304
|
checkpoint-3126/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-4689/config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 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 |
+
"output_past": true,
|
| 46 |
+
"pad_token_id": 1,
|
| 47 |
+
"position_embedding_type": "absolute",
|
| 48 |
+
"problem_type": "single_label_classification",
|
| 49 |
+
"torch_dtype": "float32",
|
| 50 |
+
"transformers_version": "4.53.3",
|
| 51 |
+
"type_vocab_size": 1,
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"vocab_size": 50265
|
| 54 |
+
}
|
checkpoint-4689/merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-4689/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93f936663e4a1ba779fe8137c5d2d3c16482ce52ce5963be2aa1d4df45cc8805
|
| 3 |
+
size 498640508
|
checkpoint-4689/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31f76a2b206c6b562d7e45aade232094212a4fa0bae8faa5a1adb47079b5cd94
|
| 3 |
+
size 997400762
|
checkpoint-4689/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e8285746cb59795639fa96739d4717f961751306375a62bc6c9c466fbeb5b4ea
|
| 3 |
+
size 14244
|
checkpoint-4689/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f58f9ddd81994851ecb68ce68d607df61fb42136fc69c1ea5d59596b18d4cc89
|
| 3 |
+
size 1064
|
checkpoint-4689/special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
checkpoint-4689/tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
checkpoint-4689/trainer_state.json
ADDED
|
@@ -0,0 +1,404 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 4689,
|
| 3 |
+
"best_metric": 0.32149304295216685,
|
| 4 |
+
"best_model_checkpoint": "/kaggle/working/codebert_b_local/checkpoint-4689",
|
| 5 |
+
"epoch": 3.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 4689,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.06397952655150352,
|
| 14 |
+
"grad_norm": 4.3835320472717285,
|
| 15 |
+
"learning_rate": 3.96e-06,
|
| 16 |
+
"loss": 1.6251,
|
| 17 |
+
"step": 100
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.12795905310300704,
|
| 21 |
+
"grad_norm": 1.2909411191940308,
|
| 22 |
+
"learning_rate": 7.960000000000002e-06,
|
| 23 |
+
"loss": 0.6209,
|
| 24 |
+
"step": 200
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.19193857965451055,
|
| 28 |
+
"grad_norm": 7.392430305480957,
|
| 29 |
+
"learning_rate": 1.196e-05,
|
| 30 |
+
"loss": 0.5322,
|
| 31 |
+
"step": 300
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.2559181062060141,
|
| 35 |
+
"grad_norm": 2.060398578643799,
|
| 36 |
+
"learning_rate": 1.5960000000000003e-05,
|
| 37 |
+
"loss": 0.4804,
|
| 38 |
+
"step": 400
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.3198976327575176,
|
| 42 |
+
"grad_norm": 19.443620681762695,
|
| 43 |
+
"learning_rate": 1.9960000000000002e-05,
|
| 44 |
+
"loss": 0.4048,
|
| 45 |
+
"step": 500
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.3838771593090211,
|
| 49 |
+
"grad_norm": 12.475482940673828,
|
| 50 |
+
"learning_rate": 1.9527333492480306e-05,
|
| 51 |
+
"loss": 0.4219,
|
| 52 |
+
"step": 600
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.44785668586052463,
|
| 56 |
+
"grad_norm": 7.762031555175781,
|
| 57 |
+
"learning_rate": 1.9049892575793746e-05,
|
| 58 |
+
"loss": 0.3789,
|
| 59 |
+
"step": 700
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.5118362124120281,
|
| 63 |
+
"grad_norm": 2.098180055618286,
|
| 64 |
+
"learning_rate": 1.8572451659107187e-05,
|
| 65 |
+
"loss": 0.4134,
|
| 66 |
+
"step": 800
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.5758157389635317,
|
| 70 |
+
"grad_norm": 2.177507162094116,
|
| 71 |
+
"learning_rate": 1.8095010742420627e-05,
|
| 72 |
+
"loss": 0.3903,
|
| 73 |
+
"step": 900
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.6397952655150352,
|
| 77 |
+
"grad_norm": 5.113386631011963,
|
| 78 |
+
"learning_rate": 1.7617569825734067e-05,
|
| 79 |
+
"loss": 0.3559,
|
| 80 |
+
"step": 1000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.7037747920665387,
|
| 84 |
+
"grad_norm": 4.909190654754639,
|
| 85 |
+
"learning_rate": 1.7140128909047508e-05,
|
| 86 |
+
"loss": 0.3219,
|
| 87 |
+
"step": 1100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.7677543186180422,
|
| 91 |
+
"grad_norm": 5.605203628540039,
|
| 92 |
+
"learning_rate": 1.6662687992360948e-05,
|
| 93 |
+
"loss": 0.3821,
|
| 94 |
+
"step": 1200
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.8317338451695457,
|
| 98 |
+
"grad_norm": 6.7283453941345215,
|
| 99 |
+
"learning_rate": 1.6185247075674385e-05,
|
| 100 |
+
"loss": 0.3674,
|
| 101 |
+
"step": 1300
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.8957133717210493,
|
| 105 |
+
"grad_norm": 4.659298419952393,
|
| 106 |
+
"learning_rate": 1.5707806158987825e-05,
|
| 107 |
+
"loss": 0.3485,
|
| 108 |
+
"step": 1400
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.9596928982725528,
|
| 112 |
+
"grad_norm": 6.4099931716918945,
|
| 113 |
+
"learning_rate": 1.5230365242301266e-05,
|
| 114 |
+
"loss": 0.3565,
|
| 115 |
+
"step": 1500
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 1.0,
|
| 119 |
+
"eval_accuracy": 0.9024,
|
| 120 |
+
"eval_f1_macro": 0.23154395234377326,
|
| 121 |
+
"eval_f1_weighted": 0.8916810274029613,
|
| 122 |
+
"eval_loss": 0.3302375078201294,
|
| 123 |
+
"eval_precision_macro": 0.2442325389892706,
|
| 124 |
+
"eval_recall_macro": 0.2618191076765516,
|
| 125 |
+
"eval_runtime": 91.0613,
|
| 126 |
+
"eval_samples_per_second": 109.816,
|
| 127 |
+
"eval_steps_per_second": 3.437,
|
| 128 |
+
"step": 1563
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.0236724248240563,
|
| 132 |
+
"grad_norm": 5.717738151550293,
|
| 133 |
+
"learning_rate": 1.4752924325614706e-05,
|
| 134 |
+
"loss": 0.318,
|
| 135 |
+
"step": 1600
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 1.0876519513755598,
|
| 139 |
+
"grad_norm": 4.969061851501465,
|
| 140 |
+
"learning_rate": 1.4275483408928146e-05,
|
| 141 |
+
"loss": 0.3001,
|
| 142 |
+
"step": 1700
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 1.1516314779270633,
|
| 146 |
+
"grad_norm": 3.3130006790161133,
|
| 147 |
+
"learning_rate": 1.3798042492241585e-05,
|
| 148 |
+
"loss": 0.3332,
|
| 149 |
+
"step": 1800
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.2156110044785668,
|
| 153 |
+
"grad_norm": 4.077260971069336,
|
| 154 |
+
"learning_rate": 1.3320601575555025e-05,
|
| 155 |
+
"loss": 0.32,
|
| 156 |
+
"step": 1900
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.2795905310300704,
|
| 160 |
+
"grad_norm": 3.1868574619293213,
|
| 161 |
+
"learning_rate": 1.2843160658868466e-05,
|
| 162 |
+
"loss": 0.3095,
|
| 163 |
+
"step": 2000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 1.3435700575815739,
|
| 167 |
+
"grad_norm": 6.247142314910889,
|
| 168 |
+
"learning_rate": 1.2365719742181906e-05,
|
| 169 |
+
"loss": 0.3105,
|
| 170 |
+
"step": 2100
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 1.4075495841330774,
|
| 174 |
+
"grad_norm": 11.029816627502441,
|
| 175 |
+
"learning_rate": 1.1888278825495345e-05,
|
| 176 |
+
"loss": 0.2975,
|
| 177 |
+
"step": 2200
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 1.471529110684581,
|
| 181 |
+
"grad_norm": 1.82353937625885,
|
| 182 |
+
"learning_rate": 1.1410837908808785e-05,
|
| 183 |
+
"loss": 0.3047,
|
| 184 |
+
"step": 2300
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.5355086372360844,
|
| 188 |
+
"grad_norm": 8.549129486083984,
|
| 189 |
+
"learning_rate": 1.0933396992122225e-05,
|
| 190 |
+
"loss": 0.2983,
|
| 191 |
+
"step": 2400
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 1.599488163787588,
|
| 195 |
+
"grad_norm": 5.218935966491699,
|
| 196 |
+
"learning_rate": 1.0455956075435664e-05,
|
| 197 |
+
"loss": 0.2965,
|
| 198 |
+
"step": 2500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 1.6634676903390915,
|
| 202 |
+
"grad_norm": 1.3732364177703857,
|
| 203 |
+
"learning_rate": 9.978515158749106e-06,
|
| 204 |
+
"loss": 0.274,
|
| 205 |
+
"step": 2600
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 1.727447216890595,
|
| 209 |
+
"grad_norm": 5.042181968688965,
|
| 210 |
+
"learning_rate": 9.501074242062547e-06,
|
| 211 |
+
"loss": 0.2804,
|
| 212 |
+
"step": 2700
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 1.7914267434420985,
|
| 216 |
+
"grad_norm": 2.0916764736175537,
|
| 217 |
+
"learning_rate": 9.023633325375985e-06,
|
| 218 |
+
"loss": 0.2689,
|
| 219 |
+
"step": 2800
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 1.855406269993602,
|
| 223 |
+
"grad_norm": 5.730693817138672,
|
| 224 |
+
"learning_rate": 8.546192408689426e-06,
|
| 225 |
+
"loss": 0.2798,
|
| 226 |
+
"step": 2900
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.9193857965451055,
|
| 230 |
+
"grad_norm": 1.510636806488037,
|
| 231 |
+
"learning_rate": 8.068751492002866e-06,
|
| 232 |
+
"loss": 0.2811,
|
| 233 |
+
"step": 3000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.983365323096609,
|
| 237 |
+
"grad_norm": 5.20271110534668,
|
| 238 |
+
"learning_rate": 7.591310575316305e-06,
|
| 239 |
+
"loss": 0.2862,
|
| 240 |
+
"step": 3100
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 2.0,
|
| 244 |
+
"eval_accuracy": 0.9119,
|
| 245 |
+
"eval_f1_macro": 0.284542217831853,
|
| 246 |
+
"eval_f1_weighted": 0.900679083795208,
|
| 247 |
+
"eval_loss": 0.30098485946655273,
|
| 248 |
+
"eval_precision_macro": 0.36539760346541894,
|
| 249 |
+
"eval_recall_macro": 0.2947437498965655,
|
| 250 |
+
"eval_runtime": 91.3907,
|
| 251 |
+
"eval_samples_per_second": 109.42,
|
| 252 |
+
"eval_steps_per_second": 3.425,
|
| 253 |
+
"step": 3126
|
| 254 |
+
},
|
| 255 |
+
{
|
| 256 |
+
"epoch": 2.0473448496481126,
|
| 257 |
+
"grad_norm": 4.349801063537598,
|
| 258 |
+
"learning_rate": 7.113869658629745e-06,
|
| 259 |
+
"loss": 0.2495,
|
| 260 |
+
"step": 3200
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 2.111324376199616,
|
| 264 |
+
"grad_norm": 4.5755085945129395,
|
| 265 |
+
"learning_rate": 6.636428741943185e-06,
|
| 266 |
+
"loss": 0.2693,
|
| 267 |
+
"step": 3300
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 2.1753039027511196,
|
| 271 |
+
"grad_norm": 5.139939308166504,
|
| 272 |
+
"learning_rate": 6.158987825256625e-06,
|
| 273 |
+
"loss": 0.2124,
|
| 274 |
+
"step": 3400
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 2.239283429302623,
|
| 278 |
+
"grad_norm": 2.8210389614105225,
|
| 279 |
+
"learning_rate": 5.681546908570065e-06,
|
| 280 |
+
"loss": 0.2504,
|
| 281 |
+
"step": 3500
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 2.3032629558541267,
|
| 285 |
+
"grad_norm": 3.741081476211548,
|
| 286 |
+
"learning_rate": 5.204105991883505e-06,
|
| 287 |
+
"loss": 0.2478,
|
| 288 |
+
"step": 3600
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 2.36724248240563,
|
| 292 |
+
"grad_norm": 5.447690010070801,
|
| 293 |
+
"learning_rate": 4.726665075196944e-06,
|
| 294 |
+
"loss": 0.2405,
|
| 295 |
+
"step": 3700
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 2.4312220089571337,
|
| 299 |
+
"grad_norm": 4.314721584320068,
|
| 300 |
+
"learning_rate": 4.249224158510385e-06,
|
| 301 |
+
"loss": 0.2561,
|
| 302 |
+
"step": 3800
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 2.495201535508637,
|
| 306 |
+
"grad_norm": 4.551843643188477,
|
| 307 |
+
"learning_rate": 3.771783241823825e-06,
|
| 308 |
+
"loss": 0.2435,
|
| 309 |
+
"step": 3900
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 2.5591810620601407,
|
| 313 |
+
"grad_norm": 3.1414506435394287,
|
| 314 |
+
"learning_rate": 3.2943423251372647e-06,
|
| 315 |
+
"loss": 0.2721,
|
| 316 |
+
"step": 4000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 2.6231605886116443,
|
| 320 |
+
"grad_norm": 8.281805992126465,
|
| 321 |
+
"learning_rate": 2.8169014084507046e-06,
|
| 322 |
+
"loss": 0.2602,
|
| 323 |
+
"step": 4100
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 2.6871401151631478,
|
| 327 |
+
"grad_norm": 4.985534191131592,
|
| 328 |
+
"learning_rate": 2.3394604917641446e-06,
|
| 329 |
+
"loss": 0.2394,
|
| 330 |
+
"step": 4200
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 2.7511196417146513,
|
| 334 |
+
"grad_norm": 7.348026275634766,
|
| 335 |
+
"learning_rate": 1.8620195750775843e-06,
|
| 336 |
+
"loss": 0.2416,
|
| 337 |
+
"step": 4300
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 2.815099168266155,
|
| 341 |
+
"grad_norm": 2.279198408126831,
|
| 342 |
+
"learning_rate": 1.3845786583910242e-06,
|
| 343 |
+
"loss": 0.2323,
|
| 344 |
+
"step": 4400
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 2.8790786948176583,
|
| 348 |
+
"grad_norm": 5.688992023468018,
|
| 349 |
+
"learning_rate": 9.071377417044641e-07,
|
| 350 |
+
"loss": 0.2455,
|
| 351 |
+
"step": 4500
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 2.943058221369162,
|
| 355 |
+
"grad_norm": 3.175658941268921,
|
| 356 |
+
"learning_rate": 4.2969682501790407e-07,
|
| 357 |
+
"loss": 0.2379,
|
| 358 |
+
"step": 4600
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.0,
|
| 362 |
+
"eval_accuracy": 0.9091,
|
| 363 |
+
"eval_f1_macro": 0.32149304295216685,
|
| 364 |
+
"eval_f1_weighted": 0.903091739400006,
|
| 365 |
+
"eval_loss": 0.305535227060318,
|
| 366 |
+
"eval_precision_macro": 0.38093123210850466,
|
| 367 |
+
"eval_recall_macro": 0.3358332333714398,
|
| 368 |
+
"eval_runtime": 91.427,
|
| 369 |
+
"eval_samples_per_second": 109.377,
|
| 370 |
+
"eval_steps_per_second": 3.423,
|
| 371 |
+
"step": 4689
|
| 372 |
+
}
|
| 373 |
+
],
|
| 374 |
+
"logging_steps": 100,
|
| 375 |
+
"max_steps": 4689,
|
| 376 |
+
"num_input_tokens_seen": 0,
|
| 377 |
+
"num_train_epochs": 3,
|
| 378 |
+
"save_steps": 500,
|
| 379 |
+
"stateful_callbacks": {
|
| 380 |
+
"EarlyStoppingCallback": {
|
| 381 |
+
"args": {
|
| 382 |
+
"early_stopping_patience": 3,
|
| 383 |
+
"early_stopping_threshold": 0.0
|
| 384 |
+
},
|
| 385 |
+
"attributes": {
|
| 386 |
+
"early_stopping_patience_counter": 0
|
| 387 |
+
}
|
| 388 |
+
},
|
| 389 |
+
"TrainerControl": {
|
| 390 |
+
"args": {
|
| 391 |
+
"should_epoch_stop": false,
|
| 392 |
+
"should_evaluate": false,
|
| 393 |
+
"should_log": false,
|
| 394 |
+
"should_save": true,
|
| 395 |
+
"should_training_stop": true
|
| 396 |
+
},
|
| 397 |
+
"attributes": {}
|
| 398 |
+
}
|
| 399 |
+
},
|
| 400 |
+
"total_flos": 1.97349237504e+16,
|
| 401 |
+
"train_batch_size": 32,
|
| 402 |
+
"trial_name": null,
|
| 403 |
+
"trial_params": null
|
| 404 |
+
}
|
checkpoint-4689/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c79d0f0062a44424d0920ddb0ddfbfcf6ff0ddf33ebfac6155bad8cb9c616ba9
|
| 3 |
+
size 5304
|
checkpoint-4689/vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,54 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"RobertaForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"bos_token_id": 0,
|
| 7 |
+
"classifier_dropout": null,
|
| 8 |
+
"eos_token_id": 2,
|
| 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 |
+
"output_past": true,
|
| 46 |
+
"pad_token_id": 1,
|
| 47 |
+
"position_embedding_type": "absolute",
|
| 48 |
+
"problem_type": "single_label_classification",
|
| 49 |
+
"torch_dtype": "float32",
|
| 50 |
+
"transformers_version": "4.53.3",
|
| 51 |
+
"type_vocab_size": 1,
|
| 52 |
+
"use_cache": true,
|
| 53 |
+
"vocab_size": 50265
|
| 54 |
+
}
|
eval_metrics.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"task": "B",
|
| 3 |
+
"num_samples": 10000,
|
| 4 |
+
"accuracy": 0.9091,
|
| 5 |
+
"f1_macro": 0.32149304295216685,
|
| 6 |
+
"f1_weighted": 0.903091739400006,
|
| 7 |
+
"precision_macro": 0.38093123210850466,
|
| 8 |
+
"recall_macro": 0.3358332333714398
|
| 9 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:93f936663e4a1ba779fe8137c5d2d3c16482ce52ce5963be2aa1d4df45cc8805
|
| 3 |
+
size 498640508
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<s>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"cls_token": {
|
| 10 |
+
"content": "<s>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"eos_token": {
|
| 17 |
+
"content": "</s>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"mask_token": {
|
| 24 |
+
"content": "<mask>",
|
| 25 |
+
"lstrip": true,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"pad_token": {
|
| 31 |
+
"content": "<pad>",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": true,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
},
|
| 37 |
+
"sep_token": {
|
| 38 |
+
"content": "</s>",
|
| 39 |
+
"lstrip": false,
|
| 40 |
+
"normalized": true,
|
| 41 |
+
"rstrip": false,
|
| 42 |
+
"single_word": false
|
| 43 |
+
},
|
| 44 |
+
"unk_token": {
|
| 45 |
+
"content": "<unk>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": true,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false
|
| 50 |
+
}
|
| 51 |
+
}
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<s>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"1": {
|
| 13 |
+
"content": "<pad>",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": true
|
| 19 |
+
},
|
| 20 |
+
"2": {
|
| 21 |
+
"content": "</s>",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": true
|
| 27 |
+
},
|
| 28 |
+
"3": {
|
| 29 |
+
"content": "<unk>",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": true
|
| 35 |
+
},
|
| 36 |
+
"50264": {
|
| 37 |
+
"content": "<mask>",
|
| 38 |
+
"lstrip": true,
|
| 39 |
+
"normalized": false,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": true
|
| 43 |
+
}
|
| 44 |
+
},
|
| 45 |
+
"bos_token": "<s>",
|
| 46 |
+
"clean_up_tokenization_spaces": false,
|
| 47 |
+
"cls_token": "<s>",
|
| 48 |
+
"eos_token": "</s>",
|
| 49 |
+
"errors": "replace",
|
| 50 |
+
"extra_special_tokens": {},
|
| 51 |
+
"mask_token": "<mask>",
|
| 52 |
+
"model_max_length": 512,
|
| 53 |
+
"pad_token": "<pad>",
|
| 54 |
+
"sep_token": "</s>",
|
| 55 |
+
"tokenizer_class": "RobertaTokenizer",
|
| 56 |
+
"unk_token": "<unk>"
|
| 57 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c79d0f0062a44424d0920ddb0ddfbfcf6ff0ddf33ebfac6155bad8cb9c616ba9
|
| 3 |
+
size 5304
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|