Training in progress, epoch 1
Browse files- .gitignore +1 -0
- config.json +40 -0
- last-checkpoint/config.json +40 -0
- last-checkpoint/optimizer.pt +3 -0
- last-checkpoint/pytorch_model.bin +3 -0
- last-checkpoint/rng_state.pth +3 -0
- last-checkpoint/scheduler.pt +3 -0
- last-checkpoint/special_tokens_map.json +15 -0
- last-checkpoint/tokenizer.json +0 -0
- last-checkpoint/tokenizer_config.json +23 -0
- last-checkpoint/trainer_state.json +812 -0
- last-checkpoint/training_args.bin +3 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +15 -0
- tokenizer.json +0 -0
- tokenizer_config.json +23 -0
- training_args.bin +3 -0
.gitignore
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
checkpoint-*/
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "albert-base-v2",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"AlbertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0,
|
| 7 |
+
"bos_token_id": 2,
|
| 8 |
+
"classifier_dropout_prob": 0.1,
|
| 9 |
+
"down_scale_factor": 1,
|
| 10 |
+
"embedding_size": 128,
|
| 11 |
+
"eos_token_id": 3,
|
| 12 |
+
"gap_size": 0,
|
| 13 |
+
"hidden_act": "gelu_new",
|
| 14 |
+
"hidden_dropout_prob": 0,
|
| 15 |
+
"hidden_size": 768,
|
| 16 |
+
"id2label": {
|
| 17 |
+
"0": "LABEL_0"
|
| 18 |
+
},
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"inner_group_num": 1,
|
| 21 |
+
"intermediate_size": 3072,
|
| 22 |
+
"label2id": {
|
| 23 |
+
"LABEL_0": 0
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "albert",
|
| 28 |
+
"net_structure_type": 0,
|
| 29 |
+
"num_attention_heads": 12,
|
| 30 |
+
"num_hidden_groups": 1,
|
| 31 |
+
"num_hidden_layers": 12,
|
| 32 |
+
"num_memory_blocks": 0,
|
| 33 |
+
"pad_token_id": 0,
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"problem_type": "regression",
|
| 36 |
+
"torch_dtype": "float32",
|
| 37 |
+
"transformers_version": "4.20.0",
|
| 38 |
+
"type_vocab_size": 2,
|
| 39 |
+
"vocab_size": 30000
|
| 40 |
+
}
|
last-checkpoint/config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "albert-base-v2",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"AlbertForSequenceClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0,
|
| 7 |
+
"bos_token_id": 2,
|
| 8 |
+
"classifier_dropout_prob": 0.1,
|
| 9 |
+
"down_scale_factor": 1,
|
| 10 |
+
"embedding_size": 128,
|
| 11 |
+
"eos_token_id": 3,
|
| 12 |
+
"gap_size": 0,
|
| 13 |
+
"hidden_act": "gelu_new",
|
| 14 |
+
"hidden_dropout_prob": 0,
|
| 15 |
+
"hidden_size": 768,
|
| 16 |
+
"id2label": {
|
| 17 |
+
"0": "LABEL_0"
|
| 18 |
+
},
|
| 19 |
+
"initializer_range": 0.02,
|
| 20 |
+
"inner_group_num": 1,
|
| 21 |
+
"intermediate_size": 3072,
|
| 22 |
+
"label2id": {
|
| 23 |
+
"LABEL_0": 0
|
| 24 |
+
},
|
| 25 |
+
"layer_norm_eps": 1e-12,
|
| 26 |
+
"max_position_embeddings": 512,
|
| 27 |
+
"model_type": "albert",
|
| 28 |
+
"net_structure_type": 0,
|
| 29 |
+
"num_attention_heads": 12,
|
| 30 |
+
"num_hidden_groups": 1,
|
| 31 |
+
"num_hidden_layers": 12,
|
| 32 |
+
"num_memory_blocks": 0,
|
| 33 |
+
"pad_token_id": 0,
|
| 34 |
+
"position_embedding_type": "absolute",
|
| 35 |
+
"problem_type": "regression",
|
| 36 |
+
"torch_dtype": "float32",
|
| 37 |
+
"transformers_version": "4.20.0",
|
| 38 |
+
"type_vocab_size": 2,
|
| 39 |
+
"vocab_size": 30000
|
| 40 |
+
}
|
last-checkpoint/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:59ee0edb754999aee3947b080a66ce9f4d06799a1dc3fa4f0e8734ec45688965
|
| 3 |
+
size 93489987
|
last-checkpoint/pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:258345eabede47051050cc6a4ac91640a43f7a41ed16f15feedb53551bfe9e52
|
| 3 |
+
size 46750353
|
last-checkpoint/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e4a36fa6df89660f509a441d9a18dcfc94af8f719cef7336c58f3ba9d6efebc0
|
| 3 |
+
size 14503
|
last-checkpoint/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e473deac3efe6957c4985f43a9f1ad176db7c890b452fb21b8ff689199ab6e23
|
| 3 |
+
size 623
|
last-checkpoint/special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"eos_token": "[SEP]",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "[MASK]",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "[SEP]",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
last-checkpoint/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
last-checkpoint/tokenizer_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": true,
|
| 5 |
+
"eos_token": "[SEP]",
|
| 6 |
+
"keep_accents": false,
|
| 7 |
+
"mask_token": {
|
| 8 |
+
"__type": "AddedToken",
|
| 9 |
+
"content": "[MASK]",
|
| 10 |
+
"lstrip": true,
|
| 11 |
+
"normalized": false,
|
| 12 |
+
"rstrip": false,
|
| 13 |
+
"single_word": false
|
| 14 |
+
},
|
| 15 |
+
"model_max_length": 512,
|
| 16 |
+
"name_or_path": "albert-base-v2",
|
| 17 |
+
"pad_token": "<pad>",
|
| 18 |
+
"remove_space": true,
|
| 19 |
+
"sep_token": "[SEP]",
|
| 20 |
+
"special_tokens_map_file": null,
|
| 21 |
+
"tokenizer_class": "AlbertTokenizer",
|
| 22 |
+
"unk_token": "<unk>"
|
| 23 |
+
}
|
last-checkpoint/trainer_state.json
ADDED
|
@@ -0,0 +1,812 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.0,
|
| 5 |
+
"global_step": 131,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 1.6e-07,
|
| 13 |
+
"loss": 0.2867,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 3.2e-07,
|
| 19 |
+
"loss": 0.2754,
|
| 20 |
+
"step": 2
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.02,
|
| 24 |
+
"learning_rate": 4.800000000000001e-07,
|
| 25 |
+
"loss": 0.2671,
|
| 26 |
+
"step": 3
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 6.4e-07,
|
| 31 |
+
"loss": 0.2218,
|
| 32 |
+
"step": 4
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 8.000000000000001e-07,
|
| 37 |
+
"loss": 0.231,
|
| 38 |
+
"step": 5
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 9.600000000000001e-07,
|
| 43 |
+
"loss": 0.2181,
|
| 44 |
+
"step": 6
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.05,
|
| 48 |
+
"learning_rate": 1.12e-06,
|
| 49 |
+
"loss": 0.1324,
|
| 50 |
+
"step": 7
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.06,
|
| 54 |
+
"learning_rate": 1.28e-06,
|
| 55 |
+
"loss": 0.2051,
|
| 56 |
+
"step": 8
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.07,
|
| 60 |
+
"learning_rate": 1.44e-06,
|
| 61 |
+
"loss": 0.2094,
|
| 62 |
+
"step": 9
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.08,
|
| 66 |
+
"learning_rate": 1.6000000000000001e-06,
|
| 67 |
+
"loss": 0.2211,
|
| 68 |
+
"step": 10
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.08,
|
| 72 |
+
"learning_rate": 1.76e-06,
|
| 73 |
+
"loss": 0.2145,
|
| 74 |
+
"step": 11
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.09,
|
| 78 |
+
"learning_rate": 1.9200000000000003e-06,
|
| 79 |
+
"loss": 0.3523,
|
| 80 |
+
"step": 12
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.1,
|
| 84 |
+
"learning_rate": 2.08e-06,
|
| 85 |
+
"loss": 0.2177,
|
| 86 |
+
"step": 13
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.11,
|
| 90 |
+
"learning_rate": 2.24e-06,
|
| 91 |
+
"loss": 0.2483,
|
| 92 |
+
"step": 14
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.11,
|
| 96 |
+
"learning_rate": 2.4000000000000003e-06,
|
| 97 |
+
"loss": 0.1413,
|
| 98 |
+
"step": 15
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.12,
|
| 102 |
+
"learning_rate": 2.56e-06,
|
| 103 |
+
"loss": 0.2232,
|
| 104 |
+
"step": 16
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.13,
|
| 108 |
+
"learning_rate": 2.7200000000000002e-06,
|
| 109 |
+
"loss": 0.137,
|
| 110 |
+
"step": 17
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.14,
|
| 114 |
+
"learning_rate": 2.88e-06,
|
| 115 |
+
"loss": 0.1971,
|
| 116 |
+
"step": 18
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.15,
|
| 120 |
+
"learning_rate": 3.04e-06,
|
| 121 |
+
"loss": 0.2301,
|
| 122 |
+
"step": 19
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.15,
|
| 126 |
+
"learning_rate": 3.2000000000000003e-06,
|
| 127 |
+
"loss": 0.2067,
|
| 128 |
+
"step": 20
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.16,
|
| 132 |
+
"learning_rate": 3.3600000000000004e-06,
|
| 133 |
+
"loss": 0.1724,
|
| 134 |
+
"step": 21
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.17,
|
| 138 |
+
"learning_rate": 3.52e-06,
|
| 139 |
+
"loss": 0.2525,
|
| 140 |
+
"step": 22
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.18,
|
| 144 |
+
"learning_rate": 3.6800000000000003e-06,
|
| 145 |
+
"loss": 0.1737,
|
| 146 |
+
"step": 23
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.18,
|
| 150 |
+
"learning_rate": 3.8400000000000005e-06,
|
| 151 |
+
"loss": 0.1791,
|
| 152 |
+
"step": 24
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.19,
|
| 156 |
+
"learning_rate": 4.000000000000001e-06,
|
| 157 |
+
"loss": 0.1425,
|
| 158 |
+
"step": 25
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.2,
|
| 162 |
+
"learning_rate": 4.16e-06,
|
| 163 |
+
"loss": 0.1442,
|
| 164 |
+
"step": 26
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.21,
|
| 168 |
+
"learning_rate": 4.32e-06,
|
| 169 |
+
"loss": 0.1717,
|
| 170 |
+
"step": 27
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.21,
|
| 174 |
+
"learning_rate": 4.48e-06,
|
| 175 |
+
"loss": 0.2364,
|
| 176 |
+
"step": 28
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.22,
|
| 180 |
+
"learning_rate": 4.6400000000000005e-06,
|
| 181 |
+
"loss": 0.1106,
|
| 182 |
+
"step": 29
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.23,
|
| 186 |
+
"learning_rate": 4.800000000000001e-06,
|
| 187 |
+
"loss": 0.1642,
|
| 188 |
+
"step": 30
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.24,
|
| 192 |
+
"learning_rate": 4.960000000000001e-06,
|
| 193 |
+
"loss": 0.1837,
|
| 194 |
+
"step": 31
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.24,
|
| 198 |
+
"learning_rate": 5.12e-06,
|
| 199 |
+
"loss": 0.1793,
|
| 200 |
+
"step": 32
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.25,
|
| 204 |
+
"learning_rate": 5.28e-06,
|
| 205 |
+
"loss": 0.148,
|
| 206 |
+
"step": 33
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.26,
|
| 210 |
+
"learning_rate": 5.4400000000000004e-06,
|
| 211 |
+
"loss": 0.107,
|
| 212 |
+
"step": 34
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.27,
|
| 216 |
+
"learning_rate": 5.600000000000001e-06,
|
| 217 |
+
"loss": 0.1271,
|
| 218 |
+
"step": 35
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.27,
|
| 222 |
+
"learning_rate": 5.76e-06,
|
| 223 |
+
"loss": 0.1509,
|
| 224 |
+
"step": 36
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.28,
|
| 228 |
+
"learning_rate": 5.92e-06,
|
| 229 |
+
"loss": 0.1718,
|
| 230 |
+
"step": 37
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.29,
|
| 234 |
+
"learning_rate": 6.08e-06,
|
| 235 |
+
"loss": 0.0919,
|
| 236 |
+
"step": 38
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.3,
|
| 240 |
+
"learning_rate": 6.24e-06,
|
| 241 |
+
"loss": 0.0946,
|
| 242 |
+
"step": 39
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.31,
|
| 246 |
+
"learning_rate": 6.4000000000000006e-06,
|
| 247 |
+
"loss": 0.0513,
|
| 248 |
+
"step": 40
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.31,
|
| 252 |
+
"learning_rate": 6.560000000000001e-06,
|
| 253 |
+
"loss": 0.0436,
|
| 254 |
+
"step": 41
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.32,
|
| 258 |
+
"learning_rate": 6.720000000000001e-06,
|
| 259 |
+
"loss": 0.0916,
|
| 260 |
+
"step": 42
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.33,
|
| 264 |
+
"learning_rate": 6.88e-06,
|
| 265 |
+
"loss": 0.0833,
|
| 266 |
+
"step": 43
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.34,
|
| 270 |
+
"learning_rate": 7.04e-06,
|
| 271 |
+
"loss": 0.1305,
|
| 272 |
+
"step": 44
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.34,
|
| 276 |
+
"learning_rate": 7.2000000000000005e-06,
|
| 277 |
+
"loss": 0.1527,
|
| 278 |
+
"step": 45
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.35,
|
| 282 |
+
"learning_rate": 7.360000000000001e-06,
|
| 283 |
+
"loss": 0.0627,
|
| 284 |
+
"step": 46
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.36,
|
| 288 |
+
"learning_rate": 7.520000000000001e-06,
|
| 289 |
+
"loss": 0.1283,
|
| 290 |
+
"step": 47
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.37,
|
| 294 |
+
"learning_rate": 7.680000000000001e-06,
|
| 295 |
+
"loss": 0.1314,
|
| 296 |
+
"step": 48
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.37,
|
| 300 |
+
"learning_rate": 7.840000000000001e-06,
|
| 301 |
+
"loss": 0.0529,
|
| 302 |
+
"step": 49
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.38,
|
| 306 |
+
"learning_rate": 8.000000000000001e-06,
|
| 307 |
+
"loss": 0.0637,
|
| 308 |
+
"step": 50
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.39,
|
| 312 |
+
"learning_rate": 8.16e-06,
|
| 313 |
+
"loss": 0.148,
|
| 314 |
+
"step": 51
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.4,
|
| 318 |
+
"learning_rate": 8.32e-06,
|
| 319 |
+
"loss": 0.1093,
|
| 320 |
+
"step": 52
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.4,
|
| 324 |
+
"learning_rate": 8.48e-06,
|
| 325 |
+
"loss": 0.0875,
|
| 326 |
+
"step": 53
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.41,
|
| 330 |
+
"learning_rate": 8.64e-06,
|
| 331 |
+
"loss": 0.1129,
|
| 332 |
+
"step": 54
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.42,
|
| 336 |
+
"learning_rate": 8.8e-06,
|
| 337 |
+
"loss": 0.0332,
|
| 338 |
+
"step": 55
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.43,
|
| 342 |
+
"learning_rate": 8.96e-06,
|
| 343 |
+
"loss": 0.1679,
|
| 344 |
+
"step": 56
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.44,
|
| 348 |
+
"learning_rate": 9.12e-06,
|
| 349 |
+
"loss": 0.0714,
|
| 350 |
+
"step": 57
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.44,
|
| 354 |
+
"learning_rate": 9.280000000000001e-06,
|
| 355 |
+
"loss": 0.0637,
|
| 356 |
+
"step": 58
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.45,
|
| 360 |
+
"learning_rate": 9.440000000000001e-06,
|
| 361 |
+
"loss": 0.0627,
|
| 362 |
+
"step": 59
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.46,
|
| 366 |
+
"learning_rate": 9.600000000000001e-06,
|
| 367 |
+
"loss": 0.1159,
|
| 368 |
+
"step": 60
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.47,
|
| 372 |
+
"learning_rate": 9.760000000000001e-06,
|
| 373 |
+
"loss": 0.1077,
|
| 374 |
+
"step": 61
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.47,
|
| 378 |
+
"learning_rate": 9.920000000000002e-06,
|
| 379 |
+
"loss": 0.0805,
|
| 380 |
+
"step": 62
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.48,
|
| 384 |
+
"learning_rate": 1.008e-05,
|
| 385 |
+
"loss": 0.0683,
|
| 386 |
+
"step": 63
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.49,
|
| 390 |
+
"learning_rate": 1.024e-05,
|
| 391 |
+
"loss": 0.061,
|
| 392 |
+
"step": 64
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.5,
|
| 396 |
+
"learning_rate": 1.04e-05,
|
| 397 |
+
"loss": 0.0835,
|
| 398 |
+
"step": 65
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.5,
|
| 402 |
+
"learning_rate": 1.056e-05,
|
| 403 |
+
"loss": 0.0322,
|
| 404 |
+
"step": 66
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.51,
|
| 408 |
+
"learning_rate": 1.072e-05,
|
| 409 |
+
"loss": 0.0551,
|
| 410 |
+
"step": 67
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.52,
|
| 414 |
+
"learning_rate": 1.0880000000000001e-05,
|
| 415 |
+
"loss": 0.1434,
|
| 416 |
+
"step": 68
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.53,
|
| 420 |
+
"learning_rate": 1.1040000000000001e-05,
|
| 421 |
+
"loss": 0.0986,
|
| 422 |
+
"step": 69
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.53,
|
| 426 |
+
"learning_rate": 1.1200000000000001e-05,
|
| 427 |
+
"loss": 0.1157,
|
| 428 |
+
"step": 70
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.54,
|
| 432 |
+
"learning_rate": 1.136e-05,
|
| 433 |
+
"loss": 0.0895,
|
| 434 |
+
"step": 71
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.55,
|
| 438 |
+
"learning_rate": 1.152e-05,
|
| 439 |
+
"loss": 0.0849,
|
| 440 |
+
"step": 72
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.56,
|
| 444 |
+
"learning_rate": 1.168e-05,
|
| 445 |
+
"loss": 0.0798,
|
| 446 |
+
"step": 73
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.56,
|
| 450 |
+
"learning_rate": 1.184e-05,
|
| 451 |
+
"loss": 0.0942,
|
| 452 |
+
"step": 74
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.57,
|
| 456 |
+
"learning_rate": 1.2e-05,
|
| 457 |
+
"loss": 0.0653,
|
| 458 |
+
"step": 75
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.58,
|
| 462 |
+
"learning_rate": 1.216e-05,
|
| 463 |
+
"loss": 0.1688,
|
| 464 |
+
"step": 76
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.59,
|
| 468 |
+
"learning_rate": 1.232e-05,
|
| 469 |
+
"loss": 0.0917,
|
| 470 |
+
"step": 77
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.6,
|
| 474 |
+
"learning_rate": 1.248e-05,
|
| 475 |
+
"loss": 0.0976,
|
| 476 |
+
"step": 78
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.6,
|
| 480 |
+
"learning_rate": 1.2640000000000001e-05,
|
| 481 |
+
"loss": 0.0685,
|
| 482 |
+
"step": 79
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.61,
|
| 486 |
+
"learning_rate": 1.2800000000000001e-05,
|
| 487 |
+
"loss": 0.1183,
|
| 488 |
+
"step": 80
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.62,
|
| 492 |
+
"learning_rate": 1.2960000000000001e-05,
|
| 493 |
+
"loss": 0.1741,
|
| 494 |
+
"step": 81
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.63,
|
| 498 |
+
"learning_rate": 1.3120000000000001e-05,
|
| 499 |
+
"loss": 0.1168,
|
| 500 |
+
"step": 82
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.63,
|
| 504 |
+
"learning_rate": 1.3280000000000002e-05,
|
| 505 |
+
"loss": 0.0252,
|
| 506 |
+
"step": 83
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.64,
|
| 510 |
+
"learning_rate": 1.3440000000000002e-05,
|
| 511 |
+
"loss": 0.0605,
|
| 512 |
+
"step": 84
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.65,
|
| 516 |
+
"learning_rate": 1.3600000000000002e-05,
|
| 517 |
+
"loss": 0.0454,
|
| 518 |
+
"step": 85
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.66,
|
| 522 |
+
"learning_rate": 1.376e-05,
|
| 523 |
+
"loss": 0.0543,
|
| 524 |
+
"step": 86
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.66,
|
| 528 |
+
"learning_rate": 1.392e-05,
|
| 529 |
+
"loss": 0.1386,
|
| 530 |
+
"step": 87
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.67,
|
| 534 |
+
"learning_rate": 1.408e-05,
|
| 535 |
+
"loss": 0.0734,
|
| 536 |
+
"step": 88
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.68,
|
| 540 |
+
"learning_rate": 1.4240000000000001e-05,
|
| 541 |
+
"loss": 0.0805,
|
| 542 |
+
"step": 89
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.69,
|
| 546 |
+
"learning_rate": 1.4400000000000001e-05,
|
| 547 |
+
"loss": 0.038,
|
| 548 |
+
"step": 90
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.69,
|
| 552 |
+
"learning_rate": 1.4560000000000001e-05,
|
| 553 |
+
"loss": 0.0851,
|
| 554 |
+
"step": 91
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.7,
|
| 558 |
+
"learning_rate": 1.4720000000000001e-05,
|
| 559 |
+
"loss": 0.1576,
|
| 560 |
+
"step": 92
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.71,
|
| 564 |
+
"learning_rate": 1.4880000000000002e-05,
|
| 565 |
+
"loss": 0.1288,
|
| 566 |
+
"step": 93
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.72,
|
| 570 |
+
"learning_rate": 1.5040000000000002e-05,
|
| 571 |
+
"loss": 0.0403,
|
| 572 |
+
"step": 94
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.73,
|
| 576 |
+
"learning_rate": 1.5200000000000002e-05,
|
| 577 |
+
"loss": 0.0861,
|
| 578 |
+
"step": 95
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.73,
|
| 582 |
+
"learning_rate": 1.5360000000000002e-05,
|
| 583 |
+
"loss": 0.068,
|
| 584 |
+
"step": 96
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.74,
|
| 588 |
+
"learning_rate": 1.552e-05,
|
| 589 |
+
"loss": 0.1086,
|
| 590 |
+
"step": 97
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.75,
|
| 594 |
+
"learning_rate": 1.5680000000000002e-05,
|
| 595 |
+
"loss": 0.0465,
|
| 596 |
+
"step": 98
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.76,
|
| 600 |
+
"learning_rate": 1.584e-05,
|
| 601 |
+
"loss": 0.0556,
|
| 602 |
+
"step": 99
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.76,
|
| 606 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 607 |
+
"loss": 0.0403,
|
| 608 |
+
"step": 100
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.77,
|
| 612 |
+
"learning_rate": 1.616e-05,
|
| 613 |
+
"loss": 0.0887,
|
| 614 |
+
"step": 101
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.78,
|
| 618 |
+
"learning_rate": 1.632e-05,
|
| 619 |
+
"loss": 0.0715,
|
| 620 |
+
"step": 102
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.79,
|
| 624 |
+
"learning_rate": 1.648e-05,
|
| 625 |
+
"loss": 0.1312,
|
| 626 |
+
"step": 103
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.79,
|
| 630 |
+
"learning_rate": 1.664e-05,
|
| 631 |
+
"loss": 0.0925,
|
| 632 |
+
"step": 104
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.8,
|
| 636 |
+
"learning_rate": 1.6800000000000002e-05,
|
| 637 |
+
"loss": 0.0816,
|
| 638 |
+
"step": 105
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.81,
|
| 642 |
+
"learning_rate": 1.696e-05,
|
| 643 |
+
"loss": 0.0841,
|
| 644 |
+
"step": 106
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.82,
|
| 648 |
+
"learning_rate": 1.7120000000000002e-05,
|
| 649 |
+
"loss": 0.1292,
|
| 650 |
+
"step": 107
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.82,
|
| 654 |
+
"learning_rate": 1.728e-05,
|
| 655 |
+
"loss": 0.0373,
|
| 656 |
+
"step": 108
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.83,
|
| 660 |
+
"learning_rate": 1.7440000000000002e-05,
|
| 661 |
+
"loss": 0.0877,
|
| 662 |
+
"step": 109
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.84,
|
| 666 |
+
"learning_rate": 1.76e-05,
|
| 667 |
+
"loss": 0.0516,
|
| 668 |
+
"step": 110
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.85,
|
| 672 |
+
"learning_rate": 1.7760000000000003e-05,
|
| 673 |
+
"loss": 0.1266,
|
| 674 |
+
"step": 111
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.85,
|
| 678 |
+
"learning_rate": 1.792e-05,
|
| 679 |
+
"loss": 0.0533,
|
| 680 |
+
"step": 112
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.86,
|
| 684 |
+
"learning_rate": 1.8080000000000003e-05,
|
| 685 |
+
"loss": 0.0739,
|
| 686 |
+
"step": 113
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.87,
|
| 690 |
+
"learning_rate": 1.824e-05,
|
| 691 |
+
"loss": 0.0975,
|
| 692 |
+
"step": 114
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.88,
|
| 696 |
+
"learning_rate": 1.8400000000000003e-05,
|
| 697 |
+
"loss": 0.0656,
|
| 698 |
+
"step": 115
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.89,
|
| 702 |
+
"learning_rate": 1.8560000000000002e-05,
|
| 703 |
+
"loss": 0.046,
|
| 704 |
+
"step": 116
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.89,
|
| 708 |
+
"learning_rate": 1.8720000000000004e-05,
|
| 709 |
+
"loss": 0.0499,
|
| 710 |
+
"step": 117
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.9,
|
| 714 |
+
"learning_rate": 1.8880000000000002e-05,
|
| 715 |
+
"loss": 0.0534,
|
| 716 |
+
"step": 118
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.91,
|
| 720 |
+
"learning_rate": 1.904e-05,
|
| 721 |
+
"loss": 0.0765,
|
| 722 |
+
"step": 119
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.92,
|
| 726 |
+
"learning_rate": 1.9200000000000003e-05,
|
| 727 |
+
"loss": 0.1537,
|
| 728 |
+
"step": 120
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.92,
|
| 732 |
+
"learning_rate": 1.936e-05,
|
| 733 |
+
"loss": 0.0759,
|
| 734 |
+
"step": 121
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.93,
|
| 738 |
+
"learning_rate": 1.9520000000000003e-05,
|
| 739 |
+
"loss": 0.082,
|
| 740 |
+
"step": 122
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.94,
|
| 744 |
+
"learning_rate": 1.968e-05,
|
| 745 |
+
"loss": 0.0541,
|
| 746 |
+
"step": 123
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.95,
|
| 750 |
+
"learning_rate": 1.9840000000000003e-05,
|
| 751 |
+
"loss": 0.0975,
|
| 752 |
+
"step": 124
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.95,
|
| 756 |
+
"learning_rate": 2e-05,
|
| 757 |
+
"loss": 0.029,
|
| 758 |
+
"step": 125
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.96,
|
| 762 |
+
"learning_rate": 1.9962264150943397e-05,
|
| 763 |
+
"loss": 0.0705,
|
| 764 |
+
"step": 126
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.97,
|
| 768 |
+
"learning_rate": 1.9924528301886793e-05,
|
| 769 |
+
"loss": 0.1127,
|
| 770 |
+
"step": 127
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.98,
|
| 774 |
+
"learning_rate": 1.988679245283019e-05,
|
| 775 |
+
"loss": 0.0691,
|
| 776 |
+
"step": 128
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.98,
|
| 780 |
+
"learning_rate": 1.9849056603773588e-05,
|
| 781 |
+
"loss": 0.0725,
|
| 782 |
+
"step": 129
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.99,
|
| 786 |
+
"learning_rate": 1.9811320754716984e-05,
|
| 787 |
+
"loss": 0.0658,
|
| 788 |
+
"step": 130
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 1.0,
|
| 792 |
+
"learning_rate": 1.977358490566038e-05,
|
| 793 |
+
"loss": 0.0294,
|
| 794 |
+
"step": 131
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 1.0,
|
| 798 |
+
"eval_loss": 0.04574168100953102,
|
| 799 |
+
"eval_pearson": 0.877006121911329,
|
| 800 |
+
"eval_runtime": 5.0274,
|
| 801 |
+
"eval_samples_per_second": 46.147,
|
| 802 |
+
"eval_spearmanr": 0.8351499431798517,
|
| 803 |
+
"eval_steps_per_second": 2.984,
|
| 804 |
+
"step": 131
|
| 805 |
+
}
|
| 806 |
+
],
|
| 807 |
+
"max_steps": 655,
|
| 808 |
+
"num_train_epochs": 5,
|
| 809 |
+
"total_flos": 49761570485592.0,
|
| 810 |
+
"trial_name": null,
|
| 811 |
+
"trial_params": null
|
| 812 |
+
}
|
last-checkpoint/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fef87647e1fed63f6ddb2c68556d0d38aa287a641f7dc0254866a55369c943a5
|
| 3 |
+
size 3311
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:258345eabede47051050cc6a4ac91640a43f7a41ed16f15feedb53551bfe9e52
|
| 3 |
+
size 46750353
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"eos_token": "[SEP]",
|
| 5 |
+
"mask_token": {
|
| 6 |
+
"content": "[MASK]",
|
| 7 |
+
"lstrip": true,
|
| 8 |
+
"normalized": false,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"pad_token": "<pad>",
|
| 13 |
+
"sep_token": "[SEP]",
|
| 14 |
+
"unk_token": "<unk>"
|
| 15 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "[CLS]",
|
| 3 |
+
"cls_token": "[CLS]",
|
| 4 |
+
"do_lower_case": true,
|
| 5 |
+
"eos_token": "[SEP]",
|
| 6 |
+
"keep_accents": false,
|
| 7 |
+
"mask_token": {
|
| 8 |
+
"__type": "AddedToken",
|
| 9 |
+
"content": "[MASK]",
|
| 10 |
+
"lstrip": true,
|
| 11 |
+
"normalized": false,
|
| 12 |
+
"rstrip": false,
|
| 13 |
+
"single_word": false
|
| 14 |
+
},
|
| 15 |
+
"model_max_length": 512,
|
| 16 |
+
"name_or_path": "albert-base-v2",
|
| 17 |
+
"pad_token": "<pad>",
|
| 18 |
+
"remove_space": true,
|
| 19 |
+
"sep_token": "[SEP]",
|
| 20 |
+
"special_tokens_map_file": null,
|
| 21 |
+
"tokenizer_class": "AlbertTokenizer",
|
| 22 |
+
"unk_token": "<unk>"
|
| 23 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:fef87647e1fed63f6ddb2c68556d0d38aa287a641f7dc0254866a55369c943a5
|
| 3 |
+
size 3311
|