Upload folder using huggingface_hub
Browse files- README.md +75 -0
- checkpoint-1500/config.json +26 -0
- checkpoint-1500/model.safetensors +3 -0
- checkpoint-1500/optimizer.pt +3 -0
- checkpoint-1500/rng_state.pth +3 -0
- checkpoint-1500/scheduler.pt +3 -0
- checkpoint-1500/special_tokens_map.json +37 -0
- checkpoint-1500/tokenizer.json +0 -0
- checkpoint-1500/tokenizer_config.json +58 -0
- checkpoint-1500/trainer_state.json +310 -0
- checkpoint-1500/training_args.bin +3 -0
- checkpoint-1500/vocab.txt +0 -0
- checkpoint-2000/config.json +26 -0
- checkpoint-2000/model.safetensors +3 -0
- checkpoint-2000/optimizer.pt +3 -0
- checkpoint-2000/rng_state.pth +3 -0
- checkpoint-2000/scheduler.pt +3 -0
- checkpoint-2000/special_tokens_map.json +37 -0
- checkpoint-2000/tokenizer.json +0 -0
- checkpoint-2000/tokenizer_config.json +58 -0
- checkpoint-2000/trainer_state.json +399 -0
- checkpoint-2000/training_args.bin +3 -0
- checkpoint-2000/vocab.txt +0 -0
- checkpoint-2280/config.json +26 -0
- checkpoint-2280/model.safetensors +3 -0
- checkpoint-2280/optimizer.pt +3 -0
- checkpoint-2280/rng_state.pth +3 -0
- checkpoint-2280/scheduler.pt +3 -0
- checkpoint-2280/special_tokens_map.json +37 -0
- checkpoint-2280/tokenizer.json +0 -0
- checkpoint-2280/tokenizer_config.json +58 -0
- checkpoint-2280/trainer_state.json +434 -0
- checkpoint-2280/training_args.bin +3 -0
- checkpoint-2280/vocab.txt +0 -0
- config.json +26 -0
- eval_metrics.json +18 -0
- model.safetensors +3 -0
- special_tokens_map.json +37 -0
- tokenizer.json +0 -0
- tokenizer_config.json +58 -0
- training_args.bin +3 -0
- vocab.txt +0 -0
README.md
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
tags:
|
| 4 |
+
- bge
|
| 5 |
+
- food
|
| 6 |
+
- reranking
|
| 7 |
+
- sequence-classification
|
| 8 |
+
- sentence-similarity
|
| 9 |
+
library_name: transformers
|
| 10 |
+
pipeline_tag: text-classification
|
| 11 |
+
license: mit
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
# Food Re-ranker Model
|
| 15 |
+
|
| 16 |
+
This is a fine-tuned BGE (BAAI General Embedding) model trained for binary classification of food description pairs. The model determines whether two food descriptions refer to the same item, enabling accurate re-ranking of search results.
|
| 17 |
+
|
| 18 |
+
## Model Details
|
| 19 |
+
|
| 20 |
+
### Model Description
|
| 21 |
+
- **Base Model**: BAAI/bge-base-en-v1.5
|
| 22 |
+
- **Model Type**: bert
|
| 23 |
+
- **Task**: Binary classification (food description matching)
|
| 24 |
+
- **Output**: Binary classification scores (0 = different foods, 1 = same food)
|
| 25 |
+
|
| 26 |
+
### Architecture
|
| 27 |
+
The model uses the BGE architecture with the following specifications:
|
| 28 |
+
- **Hidden Size**: 768
|
| 29 |
+
- **Number of Layers**: 12
|
| 30 |
+
- **Number of Attention Heads**: 12
|
| 31 |
+
- **Intermediate Size**: 3072
|
| 32 |
+
- **Maximum Position Embeddings**: 512
|
| 33 |
+
- **Vocabulary Size**: 30522
|
| 34 |
+
- **Hidden Act**: gelu
|
| 35 |
+
|
| 36 |
+
### Performance Metrics
|
| 37 |
+
Key evaluation metrics on the test set:
|
| 38 |
+
|
| 39 |
+
|
| 40 |
+
## Use Case
|
| 41 |
+
|
| 42 |
+
Designed for improving food search accuracy by re-ranking initial search results, this model:
|
| 43 |
+
- Takes pairs of food descriptions as input
|
| 44 |
+
- Determines if they refer to the same food item
|
| 45 |
+
- Enables more accurate matching of food descriptions
|
| 46 |
+
- Helps surface the most relevant matches in search results
|
| 47 |
+
|
| 48 |
+
## Training Configuration
|
| 49 |
+
- **Batch Size**: 32
|
| 50 |
+
- **Learning Rate**: 2e-05
|
| 51 |
+
- **Number of Epochs**: 10
|
| 52 |
+
- **Warmup Steps**: 0
|
| 53 |
+
- **Weight Decay**: 0.01
|
| 54 |
+
- **Dropout**: 0.1
|
| 55 |
+
- **Attention Dropout**: 0.1
|
| 56 |
+
- **Layer Norm Eps**: 1e-12
|
| 57 |
+
|
| 58 |
+
## Example Usage
|
| 59 |
+
|
| 60 |
+
```python
|
| 61 |
+
from transformers import AutoTokenizer, AutoModelForSequenceClassification
|
| 62 |
+
|
| 63 |
+
# Load model
|
| 64 |
+
tokenizer = AutoTokenizer.from_pretrained("jonny9f/food_reranker2")
|
| 65 |
+
model = AutoModelForSequenceClassification.from_pretrained("jonny9f/food_reranker2")
|
| 66 |
+
|
| 67 |
+
# Prepare input
|
| 68 |
+
query = "chicken breast"
|
| 69 |
+
candidate = "grilled chicken breast"
|
| 70 |
+
inputs = tokenizer(query, candidate, padding=True, truncation=True, return_tensors="pt")
|
| 71 |
+
|
| 72 |
+
# Get prediction
|
| 73 |
+
outputs = model(**inputs)
|
| 74 |
+
score = outputs.logits.softmax(dim=1)[0][1].item() # Score for positive class
|
| 75 |
+
```
|
checkpoint-1500/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.51.3",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
checkpoint-1500/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:85999e6c07543e7d00356ae72be27a068b634dc8ff23562e10dd2160a6e053ea
|
| 3 |
+
size 437958648
|
checkpoint-1500/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e6f96f1b0ccd7d27d9734883b475135b5db96bc60358adbb1642e2d45ec87f5f
|
| 3 |
+
size 876038795
|
checkpoint-1500/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:867ad06cb65bf1f3cd5c95e4e703f60dcd05051c966fa678f0b8e3e69b269614
|
| 3 |
+
size 14645
|
checkpoint-1500/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4246e9af2fbe4b4997361ef8d0c0da7b2e1264a2467019754c916a12a724518a
|
| 3 |
+
size 1465
|
checkpoint-1500/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-1500/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-1500/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
checkpoint-1500/trainer_state.json
ADDED
|
@@ -0,0 +1,310 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 1500,
|
| 3 |
+
"best_metric": 0.8241525423728814,
|
| 4 |
+
"best_model_checkpoint": "./models/checkpoints_v4/checkpoint-1500",
|
| 5 |
+
"epoch": 6.578947368421053,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1500,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.21929824561403508,
|
| 14 |
+
"grad_norm": 3.511173725128174,
|
| 15 |
+
"learning_rate": 4.298245614035088e-06,
|
| 16 |
+
"loss": 0.8133,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.43859649122807015,
|
| 21 |
+
"grad_norm": 5.231376647949219,
|
| 22 |
+
"learning_rate": 8.68421052631579e-06,
|
| 23 |
+
"loss": 0.5079,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.6578947368421053,
|
| 28 |
+
"grad_norm": 4.2739081382751465,
|
| 29 |
+
"learning_rate": 1.3070175438596493e-05,
|
| 30 |
+
"loss": 0.3543,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.8771929824561403,
|
| 35 |
+
"grad_norm": 4.163576602935791,
|
| 36 |
+
"learning_rate": 1.7456140350877195e-05,
|
| 37 |
+
"loss": 0.3143,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0964912280701755,
|
| 42 |
+
"grad_norm": 3.981715202331543,
|
| 43 |
+
"learning_rate": 1.9795321637426903e-05,
|
| 44 |
+
"loss": 0.3005,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.3157894736842106,
|
| 49 |
+
"grad_norm": 4.984482288360596,
|
| 50 |
+
"learning_rate": 1.9307992202729045e-05,
|
| 51 |
+
"loss": 0.2521,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.5350877192982457,
|
| 56 |
+
"grad_norm": 7.214444637298584,
|
| 57 |
+
"learning_rate": 1.882066276803119e-05,
|
| 58 |
+
"loss": 0.2481,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.7543859649122808,
|
| 63 |
+
"grad_norm": 3.6749589443206787,
|
| 64 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 65 |
+
"loss": 0.2475,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.973684210526316,
|
| 70 |
+
"grad_norm": 5.445246696472168,
|
| 71 |
+
"learning_rate": 1.784600389863548e-05,
|
| 72 |
+
"loss": 0.2289,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.192982456140351,
|
| 77 |
+
"grad_norm": 12.260554313659668,
|
| 78 |
+
"learning_rate": 1.7358674463937624e-05,
|
| 79 |
+
"loss": 0.1811,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.192982456140351,
|
| 84 |
+
"eval_accuracy": 0.8987154150197628,
|
| 85 |
+
"eval_auc": 0.9507858176268903,
|
| 86 |
+
"eval_f1": 0.7724750277469479,
|
| 87 |
+
"eval_false_negatives": 147,
|
| 88 |
+
"eval_false_positives": 58,
|
| 89 |
+
"eval_loss": 0.29392391443252563,
|
| 90 |
+
"eval_npv": 0.9091470951792336,
|
| 91 |
+
"eval_precision": 0.8571428571428571,
|
| 92 |
+
"eval_recall": 0.703030303030303,
|
| 93 |
+
"eval_runtime": 7.4602,
|
| 94 |
+
"eval_samples_per_second": 271.308,
|
| 95 |
+
"eval_specificity": 0.9620667102681492,
|
| 96 |
+
"eval_steps_per_second": 4.289,
|
| 97 |
+
"eval_true_negatives": 1471,
|
| 98 |
+
"eval_true_positives": 348,
|
| 99 |
+
"step": 500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.412280701754386,
|
| 103 |
+
"grad_norm": 3.9946436882019043,
|
| 104 |
+
"learning_rate": 1.6871345029239766e-05,
|
| 105 |
+
"loss": 0.1934,
|
| 106 |
+
"step": 550
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6315789473684212,
|
| 110 |
+
"grad_norm": 6.733097553253174,
|
| 111 |
+
"learning_rate": 1.638401559454191e-05,
|
| 112 |
+
"loss": 0.1759,
|
| 113 |
+
"step": 600
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.8508771929824563,
|
| 117 |
+
"grad_norm": 12.989212036132812,
|
| 118 |
+
"learning_rate": 1.5896686159844057e-05,
|
| 119 |
+
"loss": 0.1953,
|
| 120 |
+
"step": 650
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.0701754385964914,
|
| 124 |
+
"grad_norm": 4.068276882171631,
|
| 125 |
+
"learning_rate": 1.5409356725146202e-05,
|
| 126 |
+
"loss": 0.1248,
|
| 127 |
+
"step": 700
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.2894736842105265,
|
| 131 |
+
"grad_norm": 0.6126830577850342,
|
| 132 |
+
"learning_rate": 1.4922027290448344e-05,
|
| 133 |
+
"loss": 0.1289,
|
| 134 |
+
"step": 750
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.5087719298245617,
|
| 138 |
+
"grad_norm": 6.3761796951293945,
|
| 139 |
+
"learning_rate": 1.4434697855750488e-05,
|
| 140 |
+
"loss": 0.1117,
|
| 141 |
+
"step": 800
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.7280701754385968,
|
| 145 |
+
"grad_norm": 4.546126365661621,
|
| 146 |
+
"learning_rate": 1.3947368421052631e-05,
|
| 147 |
+
"loss": 0.1491,
|
| 148 |
+
"step": 850
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.9473684210526314,
|
| 152 |
+
"grad_norm": 4.006205081939697,
|
| 153 |
+
"learning_rate": 1.3460038986354777e-05,
|
| 154 |
+
"loss": 0.1345,
|
| 155 |
+
"step": 900
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 4.166666666666667,
|
| 159 |
+
"grad_norm": 2.2420225143432617,
|
| 160 |
+
"learning_rate": 1.297270955165692e-05,
|
| 161 |
+
"loss": 0.0904,
|
| 162 |
+
"step": 950
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 4.385964912280702,
|
| 166 |
+
"grad_norm": 3.2088565826416016,
|
| 167 |
+
"learning_rate": 1.2485380116959064e-05,
|
| 168 |
+
"loss": 0.0996,
|
| 169 |
+
"step": 1000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.385964912280702,
|
| 173 |
+
"eval_accuracy": 0.9115612648221344,
|
| 174 |
+
"eval_auc": 0.9578519002979434,
|
| 175 |
+
"eval_f1": 0.8182741116751269,
|
| 176 |
+
"eval_false_negatives": 92,
|
| 177 |
+
"eval_false_positives": 87,
|
| 178 |
+
"eval_loss": 0.3153094947338104,
|
| 179 |
+
"eval_npv": 0.940026075619296,
|
| 180 |
+
"eval_precision": 0.8224489795918367,
|
| 181 |
+
"eval_recall": 0.8141414141414142,
|
| 182 |
+
"eval_runtime": 7.4628,
|
| 183 |
+
"eval_samples_per_second": 271.212,
|
| 184 |
+
"eval_specificity": 0.9431000654022237,
|
| 185 |
+
"eval_steps_per_second": 4.288,
|
| 186 |
+
"eval_true_negatives": 1442,
|
| 187 |
+
"eval_true_positives": 403,
|
| 188 |
+
"step": 1000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.605263157894737,
|
| 192 |
+
"grad_norm": 10.743696212768555,
|
| 193 |
+
"learning_rate": 1.1998050682261208e-05,
|
| 194 |
+
"loss": 0.0909,
|
| 195 |
+
"step": 1050
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.824561403508772,
|
| 199 |
+
"grad_norm": 1.6505861282348633,
|
| 200 |
+
"learning_rate": 1.1510721247563355e-05,
|
| 201 |
+
"loss": 0.088,
|
| 202 |
+
"step": 1100
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 5.043859649122807,
|
| 206 |
+
"grad_norm": 6.567928314208984,
|
| 207 |
+
"learning_rate": 1.1023391812865499e-05,
|
| 208 |
+
"loss": 0.1129,
|
| 209 |
+
"step": 1150
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.2631578947368425,
|
| 213 |
+
"grad_norm": 3.445005416870117,
|
| 214 |
+
"learning_rate": 1.0536062378167643e-05,
|
| 215 |
+
"loss": 0.0605,
|
| 216 |
+
"step": 1200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 5.482456140350878,
|
| 220 |
+
"grad_norm": 1.5516074895858765,
|
| 221 |
+
"learning_rate": 1.0048732943469786e-05,
|
| 222 |
+
"loss": 0.0642,
|
| 223 |
+
"step": 1250
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 5.701754385964913,
|
| 227 |
+
"grad_norm": 9.6685209274292,
|
| 228 |
+
"learning_rate": 9.56140350877193e-06,
|
| 229 |
+
"loss": 0.0759,
|
| 230 |
+
"step": 1300
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 5.921052631578947,
|
| 234 |
+
"grad_norm": 12.226219177246094,
|
| 235 |
+
"learning_rate": 9.074074074074075e-06,
|
| 236 |
+
"loss": 0.0806,
|
| 237 |
+
"step": 1350
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 6.140350877192983,
|
| 241 |
+
"grad_norm": 1.530657172203064,
|
| 242 |
+
"learning_rate": 8.586744639376219e-06,
|
| 243 |
+
"loss": 0.0636,
|
| 244 |
+
"step": 1400
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 6.359649122807017,
|
| 248 |
+
"grad_norm": 4.697056293487549,
|
| 249 |
+
"learning_rate": 8.099415204678363e-06,
|
| 250 |
+
"loss": 0.0402,
|
| 251 |
+
"step": 1450
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 6.578947368421053,
|
| 255 |
+
"grad_norm": 0.09656322002410889,
|
| 256 |
+
"learning_rate": 7.612085769980507e-06,
|
| 257 |
+
"loss": 0.0629,
|
| 258 |
+
"step": 1500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 6.578947368421053,
|
| 262 |
+
"eval_accuracy": 0.9179841897233202,
|
| 263 |
+
"eval_auc": 0.9618949468524354,
|
| 264 |
+
"eval_f1": 0.8241525423728814,
|
| 265 |
+
"eval_false_negatives": 106,
|
| 266 |
+
"eval_false_positives": 60,
|
| 267 |
+
"eval_loss": 0.38592642545700073,
|
| 268 |
+
"eval_npv": 0.9326984126984127,
|
| 269 |
+
"eval_precision": 0.8663697104677061,
|
| 270 |
+
"eval_recall": 0.7858585858585858,
|
| 271 |
+
"eval_runtime": 7.4828,
|
| 272 |
+
"eval_samples_per_second": 270.486,
|
| 273 |
+
"eval_specificity": 0.960758665794637,
|
| 274 |
+
"eval_steps_per_second": 4.276,
|
| 275 |
+
"eval_true_negatives": 1469,
|
| 276 |
+
"eval_true_positives": 389,
|
| 277 |
+
"step": 1500
|
| 278 |
+
}
|
| 279 |
+
],
|
| 280 |
+
"logging_steps": 50,
|
| 281 |
+
"max_steps": 2280,
|
| 282 |
+
"num_input_tokens_seen": 0,
|
| 283 |
+
"num_train_epochs": 10,
|
| 284 |
+
"save_steps": 500,
|
| 285 |
+
"stateful_callbacks": {
|
| 286 |
+
"EarlyStoppingCallback": {
|
| 287 |
+
"args": {
|
| 288 |
+
"early_stopping_patience": 3,
|
| 289 |
+
"early_stopping_threshold": 0.001
|
| 290 |
+
},
|
| 291 |
+
"attributes": {
|
| 292 |
+
"early_stopping_patience_counter": 0
|
| 293 |
+
}
|
| 294 |
+
},
|
| 295 |
+
"TrainerControl": {
|
| 296 |
+
"args": {
|
| 297 |
+
"should_epoch_stop": false,
|
| 298 |
+
"should_evaluate": false,
|
| 299 |
+
"should_log": false,
|
| 300 |
+
"should_save": true,
|
| 301 |
+
"should_training_stop": false
|
| 302 |
+
},
|
| 303 |
+
"attributes": {}
|
| 304 |
+
}
|
| 305 |
+
},
|
| 306 |
+
"total_flos": 6303614664314880.0,
|
| 307 |
+
"train_batch_size": 32,
|
| 308 |
+
"trial_name": null,
|
| 309 |
+
"trial_params": null
|
| 310 |
+
}
|
checkpoint-1500/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fa0b6e71cf97dcf125328dcde4d9907006c97167c4f41af8ac98ea5f88ac6b4
|
| 3 |
+
size 5713
|
checkpoint-1500/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.51.3",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
checkpoint-2000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41590102e903624f664df64252a13ee88e911b1cb5be074c17b4957d02712d72
|
| 3 |
+
size 437958648
|
checkpoint-2000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9540b36b35a13d5702f7e16c9325860ffefc231e7ff5bcfd89f94e7f795172fe
|
| 3 |
+
size 876038795
|
checkpoint-2000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9f129fc5e5d59c080e9fd48b5d303d31b63156f53c9c6d279333ff06b06681a
|
| 3 |
+
size 14645
|
checkpoint-2000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:5fab9a9e4e1a8e90792d09d3ab98391d1f33f696a4645f21acb71aadc7c4ebe0
|
| 3 |
+
size 1465
|
checkpoint-2000/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-2000/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
checkpoint-2000/trainer_state.json
ADDED
|
@@ -0,0 +1,399 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2000,
|
| 3 |
+
"best_metric": 0.8264462809917356,
|
| 4 |
+
"best_model_checkpoint": "./models/checkpoints_v4/checkpoint-2000",
|
| 5 |
+
"epoch": 8.771929824561404,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2000,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.21929824561403508,
|
| 14 |
+
"grad_norm": 3.511173725128174,
|
| 15 |
+
"learning_rate": 4.298245614035088e-06,
|
| 16 |
+
"loss": 0.8133,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.43859649122807015,
|
| 21 |
+
"grad_norm": 5.231376647949219,
|
| 22 |
+
"learning_rate": 8.68421052631579e-06,
|
| 23 |
+
"loss": 0.5079,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.6578947368421053,
|
| 28 |
+
"grad_norm": 4.2739081382751465,
|
| 29 |
+
"learning_rate": 1.3070175438596493e-05,
|
| 30 |
+
"loss": 0.3543,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.8771929824561403,
|
| 35 |
+
"grad_norm": 4.163576602935791,
|
| 36 |
+
"learning_rate": 1.7456140350877195e-05,
|
| 37 |
+
"loss": 0.3143,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0964912280701755,
|
| 42 |
+
"grad_norm": 3.981715202331543,
|
| 43 |
+
"learning_rate": 1.9795321637426903e-05,
|
| 44 |
+
"loss": 0.3005,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.3157894736842106,
|
| 49 |
+
"grad_norm": 4.984482288360596,
|
| 50 |
+
"learning_rate": 1.9307992202729045e-05,
|
| 51 |
+
"loss": 0.2521,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.5350877192982457,
|
| 56 |
+
"grad_norm": 7.214444637298584,
|
| 57 |
+
"learning_rate": 1.882066276803119e-05,
|
| 58 |
+
"loss": 0.2481,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.7543859649122808,
|
| 63 |
+
"grad_norm": 3.6749589443206787,
|
| 64 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 65 |
+
"loss": 0.2475,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.973684210526316,
|
| 70 |
+
"grad_norm": 5.445246696472168,
|
| 71 |
+
"learning_rate": 1.784600389863548e-05,
|
| 72 |
+
"loss": 0.2289,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.192982456140351,
|
| 77 |
+
"grad_norm": 12.260554313659668,
|
| 78 |
+
"learning_rate": 1.7358674463937624e-05,
|
| 79 |
+
"loss": 0.1811,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.192982456140351,
|
| 84 |
+
"eval_accuracy": 0.8987154150197628,
|
| 85 |
+
"eval_auc": 0.9507858176268903,
|
| 86 |
+
"eval_f1": 0.7724750277469479,
|
| 87 |
+
"eval_false_negatives": 147,
|
| 88 |
+
"eval_false_positives": 58,
|
| 89 |
+
"eval_loss": 0.29392391443252563,
|
| 90 |
+
"eval_npv": 0.9091470951792336,
|
| 91 |
+
"eval_precision": 0.8571428571428571,
|
| 92 |
+
"eval_recall": 0.703030303030303,
|
| 93 |
+
"eval_runtime": 7.4602,
|
| 94 |
+
"eval_samples_per_second": 271.308,
|
| 95 |
+
"eval_specificity": 0.9620667102681492,
|
| 96 |
+
"eval_steps_per_second": 4.289,
|
| 97 |
+
"eval_true_negatives": 1471,
|
| 98 |
+
"eval_true_positives": 348,
|
| 99 |
+
"step": 500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.412280701754386,
|
| 103 |
+
"grad_norm": 3.9946436882019043,
|
| 104 |
+
"learning_rate": 1.6871345029239766e-05,
|
| 105 |
+
"loss": 0.1934,
|
| 106 |
+
"step": 550
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6315789473684212,
|
| 110 |
+
"grad_norm": 6.733097553253174,
|
| 111 |
+
"learning_rate": 1.638401559454191e-05,
|
| 112 |
+
"loss": 0.1759,
|
| 113 |
+
"step": 600
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.8508771929824563,
|
| 117 |
+
"grad_norm": 12.989212036132812,
|
| 118 |
+
"learning_rate": 1.5896686159844057e-05,
|
| 119 |
+
"loss": 0.1953,
|
| 120 |
+
"step": 650
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.0701754385964914,
|
| 124 |
+
"grad_norm": 4.068276882171631,
|
| 125 |
+
"learning_rate": 1.5409356725146202e-05,
|
| 126 |
+
"loss": 0.1248,
|
| 127 |
+
"step": 700
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.2894736842105265,
|
| 131 |
+
"grad_norm": 0.6126830577850342,
|
| 132 |
+
"learning_rate": 1.4922027290448344e-05,
|
| 133 |
+
"loss": 0.1289,
|
| 134 |
+
"step": 750
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.5087719298245617,
|
| 138 |
+
"grad_norm": 6.3761796951293945,
|
| 139 |
+
"learning_rate": 1.4434697855750488e-05,
|
| 140 |
+
"loss": 0.1117,
|
| 141 |
+
"step": 800
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.7280701754385968,
|
| 145 |
+
"grad_norm": 4.546126365661621,
|
| 146 |
+
"learning_rate": 1.3947368421052631e-05,
|
| 147 |
+
"loss": 0.1491,
|
| 148 |
+
"step": 850
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.9473684210526314,
|
| 152 |
+
"grad_norm": 4.006205081939697,
|
| 153 |
+
"learning_rate": 1.3460038986354777e-05,
|
| 154 |
+
"loss": 0.1345,
|
| 155 |
+
"step": 900
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 4.166666666666667,
|
| 159 |
+
"grad_norm": 2.2420225143432617,
|
| 160 |
+
"learning_rate": 1.297270955165692e-05,
|
| 161 |
+
"loss": 0.0904,
|
| 162 |
+
"step": 950
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 4.385964912280702,
|
| 166 |
+
"grad_norm": 3.2088565826416016,
|
| 167 |
+
"learning_rate": 1.2485380116959064e-05,
|
| 168 |
+
"loss": 0.0996,
|
| 169 |
+
"step": 1000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.385964912280702,
|
| 173 |
+
"eval_accuracy": 0.9115612648221344,
|
| 174 |
+
"eval_auc": 0.9578519002979434,
|
| 175 |
+
"eval_f1": 0.8182741116751269,
|
| 176 |
+
"eval_false_negatives": 92,
|
| 177 |
+
"eval_false_positives": 87,
|
| 178 |
+
"eval_loss": 0.3153094947338104,
|
| 179 |
+
"eval_npv": 0.940026075619296,
|
| 180 |
+
"eval_precision": 0.8224489795918367,
|
| 181 |
+
"eval_recall": 0.8141414141414142,
|
| 182 |
+
"eval_runtime": 7.4628,
|
| 183 |
+
"eval_samples_per_second": 271.212,
|
| 184 |
+
"eval_specificity": 0.9431000654022237,
|
| 185 |
+
"eval_steps_per_second": 4.288,
|
| 186 |
+
"eval_true_negatives": 1442,
|
| 187 |
+
"eval_true_positives": 403,
|
| 188 |
+
"step": 1000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.605263157894737,
|
| 192 |
+
"grad_norm": 10.743696212768555,
|
| 193 |
+
"learning_rate": 1.1998050682261208e-05,
|
| 194 |
+
"loss": 0.0909,
|
| 195 |
+
"step": 1050
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.824561403508772,
|
| 199 |
+
"grad_norm": 1.6505861282348633,
|
| 200 |
+
"learning_rate": 1.1510721247563355e-05,
|
| 201 |
+
"loss": 0.088,
|
| 202 |
+
"step": 1100
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 5.043859649122807,
|
| 206 |
+
"grad_norm": 6.567928314208984,
|
| 207 |
+
"learning_rate": 1.1023391812865499e-05,
|
| 208 |
+
"loss": 0.1129,
|
| 209 |
+
"step": 1150
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.2631578947368425,
|
| 213 |
+
"grad_norm": 3.445005416870117,
|
| 214 |
+
"learning_rate": 1.0536062378167643e-05,
|
| 215 |
+
"loss": 0.0605,
|
| 216 |
+
"step": 1200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 5.482456140350878,
|
| 220 |
+
"grad_norm": 1.5516074895858765,
|
| 221 |
+
"learning_rate": 1.0048732943469786e-05,
|
| 222 |
+
"loss": 0.0642,
|
| 223 |
+
"step": 1250
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 5.701754385964913,
|
| 227 |
+
"grad_norm": 9.6685209274292,
|
| 228 |
+
"learning_rate": 9.56140350877193e-06,
|
| 229 |
+
"loss": 0.0759,
|
| 230 |
+
"step": 1300
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 5.921052631578947,
|
| 234 |
+
"grad_norm": 12.226219177246094,
|
| 235 |
+
"learning_rate": 9.074074074074075e-06,
|
| 236 |
+
"loss": 0.0806,
|
| 237 |
+
"step": 1350
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 6.140350877192983,
|
| 241 |
+
"grad_norm": 1.530657172203064,
|
| 242 |
+
"learning_rate": 8.586744639376219e-06,
|
| 243 |
+
"loss": 0.0636,
|
| 244 |
+
"step": 1400
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 6.359649122807017,
|
| 248 |
+
"grad_norm": 4.697056293487549,
|
| 249 |
+
"learning_rate": 8.099415204678363e-06,
|
| 250 |
+
"loss": 0.0402,
|
| 251 |
+
"step": 1450
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 6.578947368421053,
|
| 255 |
+
"grad_norm": 0.09656322002410889,
|
| 256 |
+
"learning_rate": 7.612085769980507e-06,
|
| 257 |
+
"loss": 0.0629,
|
| 258 |
+
"step": 1500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 6.578947368421053,
|
| 262 |
+
"eval_accuracy": 0.9179841897233202,
|
| 263 |
+
"eval_auc": 0.9618949468524354,
|
| 264 |
+
"eval_f1": 0.8241525423728814,
|
| 265 |
+
"eval_false_negatives": 106,
|
| 266 |
+
"eval_false_positives": 60,
|
| 267 |
+
"eval_loss": 0.38592642545700073,
|
| 268 |
+
"eval_npv": 0.9326984126984127,
|
| 269 |
+
"eval_precision": 0.8663697104677061,
|
| 270 |
+
"eval_recall": 0.7858585858585858,
|
| 271 |
+
"eval_runtime": 7.4828,
|
| 272 |
+
"eval_samples_per_second": 270.486,
|
| 273 |
+
"eval_specificity": 0.960758665794637,
|
| 274 |
+
"eval_steps_per_second": 4.276,
|
| 275 |
+
"eval_true_negatives": 1469,
|
| 276 |
+
"eval_true_positives": 389,
|
| 277 |
+
"step": 1500
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.798245614035087,
|
| 281 |
+
"grad_norm": 36.65667724609375,
|
| 282 |
+
"learning_rate": 7.124756335282652e-06,
|
| 283 |
+
"loss": 0.0606,
|
| 284 |
+
"step": 1550
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 7.017543859649122,
|
| 288 |
+
"grad_norm": 2.3685500621795654,
|
| 289 |
+
"learning_rate": 6.637426900584796e-06,
|
| 290 |
+
"loss": 0.0513,
|
| 291 |
+
"step": 1600
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 7.2368421052631575,
|
| 295 |
+
"grad_norm": 2.3388023376464844,
|
| 296 |
+
"learning_rate": 6.15009746588694e-06,
|
| 297 |
+
"loss": 0.0328,
|
| 298 |
+
"step": 1650
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 7.456140350877193,
|
| 302 |
+
"grad_norm": 0.18014875054359436,
|
| 303 |
+
"learning_rate": 5.662768031189084e-06,
|
| 304 |
+
"loss": 0.0526,
|
| 305 |
+
"step": 1700
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 7.675438596491228,
|
| 309 |
+
"grad_norm": 5.354085445404053,
|
| 310 |
+
"learning_rate": 5.175438596491229e-06,
|
| 311 |
+
"loss": 0.033,
|
| 312 |
+
"step": 1750
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 7.894736842105263,
|
| 316 |
+
"grad_norm": 0.14219924807548523,
|
| 317 |
+
"learning_rate": 4.688109161793373e-06,
|
| 318 |
+
"loss": 0.0286,
|
| 319 |
+
"step": 1800
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.114035087719298,
|
| 323 |
+
"grad_norm": 0.8848229646682739,
|
| 324 |
+
"learning_rate": 4.200779727095517e-06,
|
| 325 |
+
"loss": 0.023,
|
| 326 |
+
"step": 1850
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.333333333333334,
|
| 330 |
+
"grad_norm": 0.09083296358585358,
|
| 331 |
+
"learning_rate": 3.713450292397661e-06,
|
| 332 |
+
"loss": 0.0342,
|
| 333 |
+
"step": 1900
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.552631578947368,
|
| 337 |
+
"grad_norm": 5.818343162536621,
|
| 338 |
+
"learning_rate": 3.2261208576998056e-06,
|
| 339 |
+
"loss": 0.0317,
|
| 340 |
+
"step": 1950
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 8.771929824561404,
|
| 344 |
+
"grad_norm": 0.027245348319411278,
|
| 345 |
+
"learning_rate": 2.7387914230019497e-06,
|
| 346 |
+
"loss": 0.0344,
|
| 347 |
+
"step": 2000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.771929824561404,
|
| 351 |
+
"eval_accuracy": 0.9169960474308301,
|
| 352 |
+
"eval_auc": 0.9587609251441821,
|
| 353 |
+
"eval_f1": 0.8264462809917356,
|
| 354 |
+
"eval_false_negatives": 95,
|
| 355 |
+
"eval_false_positives": 73,
|
| 356 |
+
"eval_loss": 0.44914552569389343,
|
| 357 |
+
"eval_npv": 0.938749194068343,
|
| 358 |
+
"eval_precision": 0.8456659619450317,
|
| 359 |
+
"eval_recall": 0.8080808080808081,
|
| 360 |
+
"eval_runtime": 7.4877,
|
| 361 |
+
"eval_samples_per_second": 270.31,
|
| 362 |
+
"eval_specificity": 0.9522563767168084,
|
| 363 |
+
"eval_steps_per_second": 4.274,
|
| 364 |
+
"eval_true_negatives": 1456,
|
| 365 |
+
"eval_true_positives": 400,
|
| 366 |
+
"step": 2000
|
| 367 |
+
}
|
| 368 |
+
],
|
| 369 |
+
"logging_steps": 50,
|
| 370 |
+
"max_steps": 2280,
|
| 371 |
+
"num_input_tokens_seen": 0,
|
| 372 |
+
"num_train_epochs": 10,
|
| 373 |
+
"save_steps": 500,
|
| 374 |
+
"stateful_callbacks": {
|
| 375 |
+
"EarlyStoppingCallback": {
|
| 376 |
+
"args": {
|
| 377 |
+
"early_stopping_patience": 3,
|
| 378 |
+
"early_stopping_threshold": 0.001
|
| 379 |
+
},
|
| 380 |
+
"attributes": {
|
| 381 |
+
"early_stopping_patience_counter": 0
|
| 382 |
+
}
|
| 383 |
+
},
|
| 384 |
+
"TrainerControl": {
|
| 385 |
+
"args": {
|
| 386 |
+
"should_epoch_stop": false,
|
| 387 |
+
"should_evaluate": false,
|
| 388 |
+
"should_log": false,
|
| 389 |
+
"should_save": true,
|
| 390 |
+
"should_training_stop": false
|
| 391 |
+
},
|
| 392 |
+
"attributes": {}
|
| 393 |
+
}
|
| 394 |
+
},
|
| 395 |
+
"total_flos": 8404819552419840.0,
|
| 396 |
+
"train_batch_size": 32,
|
| 397 |
+
"trial_name": null,
|
| 398 |
+
"trial_params": null
|
| 399 |
+
}
|
checkpoint-2000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fa0b6e71cf97dcf125328dcde4d9907006c97167c4f41af8ac98ea5f88ac6b4
|
| 3 |
+
size 5713
|
checkpoint-2000/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2280/config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.51.3",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
checkpoint-2280/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:898191b3113c997f08feb51d48ffff8478f23c749d621d6ec03b43acea202416
|
| 3 |
+
size 437958648
|
checkpoint-2280/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec925b71bf77e7f6c8815d63a7b3484f3b38ba069d09ccc30e4a15db317c4cdb
|
| 3 |
+
size 876038795
|
checkpoint-2280/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1cd6b56f462107b8bd25375a943da9b9dfe7f453a3380ae0c6204366bc7628a0
|
| 3 |
+
size 14645
|
checkpoint-2280/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:319d0be530cda2200e1d0a0824971ef0e84d372efc641975cd1f3d064cf6b46a
|
| 3 |
+
size 1465
|
checkpoint-2280/special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
checkpoint-2280/tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
checkpoint-2280/tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
checkpoint-2280/trainer_state.json
ADDED
|
@@ -0,0 +1,434 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 2000,
|
| 3 |
+
"best_metric": 0.8264462809917356,
|
| 4 |
+
"best_model_checkpoint": "./models/checkpoints_v4/checkpoint-2000",
|
| 5 |
+
"epoch": 10.0,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 2280,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.21929824561403508,
|
| 14 |
+
"grad_norm": 3.511173725128174,
|
| 15 |
+
"learning_rate": 4.298245614035088e-06,
|
| 16 |
+
"loss": 0.8133,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.43859649122807015,
|
| 21 |
+
"grad_norm": 5.231376647949219,
|
| 22 |
+
"learning_rate": 8.68421052631579e-06,
|
| 23 |
+
"loss": 0.5079,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.6578947368421053,
|
| 28 |
+
"grad_norm": 4.2739081382751465,
|
| 29 |
+
"learning_rate": 1.3070175438596493e-05,
|
| 30 |
+
"loss": 0.3543,
|
| 31 |
+
"step": 150
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.8771929824561403,
|
| 35 |
+
"grad_norm": 4.163576602935791,
|
| 36 |
+
"learning_rate": 1.7456140350877195e-05,
|
| 37 |
+
"loss": 0.3143,
|
| 38 |
+
"step": 200
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 1.0964912280701755,
|
| 42 |
+
"grad_norm": 3.981715202331543,
|
| 43 |
+
"learning_rate": 1.9795321637426903e-05,
|
| 44 |
+
"loss": 0.3005,
|
| 45 |
+
"step": 250
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 1.3157894736842106,
|
| 49 |
+
"grad_norm": 4.984482288360596,
|
| 50 |
+
"learning_rate": 1.9307992202729045e-05,
|
| 51 |
+
"loss": 0.2521,
|
| 52 |
+
"step": 300
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 1.5350877192982457,
|
| 56 |
+
"grad_norm": 7.214444637298584,
|
| 57 |
+
"learning_rate": 1.882066276803119e-05,
|
| 58 |
+
"loss": 0.2481,
|
| 59 |
+
"step": 350
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.7543859649122808,
|
| 63 |
+
"grad_norm": 3.6749589443206787,
|
| 64 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 65 |
+
"loss": 0.2475,
|
| 66 |
+
"step": 400
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.973684210526316,
|
| 70 |
+
"grad_norm": 5.445246696472168,
|
| 71 |
+
"learning_rate": 1.784600389863548e-05,
|
| 72 |
+
"loss": 0.2289,
|
| 73 |
+
"step": 450
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.192982456140351,
|
| 77 |
+
"grad_norm": 12.260554313659668,
|
| 78 |
+
"learning_rate": 1.7358674463937624e-05,
|
| 79 |
+
"loss": 0.1811,
|
| 80 |
+
"step": 500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.192982456140351,
|
| 84 |
+
"eval_accuracy": 0.8987154150197628,
|
| 85 |
+
"eval_auc": 0.9507858176268903,
|
| 86 |
+
"eval_f1": 0.7724750277469479,
|
| 87 |
+
"eval_false_negatives": 147,
|
| 88 |
+
"eval_false_positives": 58,
|
| 89 |
+
"eval_loss": 0.29392391443252563,
|
| 90 |
+
"eval_npv": 0.9091470951792336,
|
| 91 |
+
"eval_precision": 0.8571428571428571,
|
| 92 |
+
"eval_recall": 0.703030303030303,
|
| 93 |
+
"eval_runtime": 7.4602,
|
| 94 |
+
"eval_samples_per_second": 271.308,
|
| 95 |
+
"eval_specificity": 0.9620667102681492,
|
| 96 |
+
"eval_steps_per_second": 4.289,
|
| 97 |
+
"eval_true_negatives": 1471,
|
| 98 |
+
"eval_true_positives": 348,
|
| 99 |
+
"step": 500
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.412280701754386,
|
| 103 |
+
"grad_norm": 3.9946436882019043,
|
| 104 |
+
"learning_rate": 1.6871345029239766e-05,
|
| 105 |
+
"loss": 0.1934,
|
| 106 |
+
"step": 550
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 2.6315789473684212,
|
| 110 |
+
"grad_norm": 6.733097553253174,
|
| 111 |
+
"learning_rate": 1.638401559454191e-05,
|
| 112 |
+
"loss": 0.1759,
|
| 113 |
+
"step": 600
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.8508771929824563,
|
| 117 |
+
"grad_norm": 12.989212036132812,
|
| 118 |
+
"learning_rate": 1.5896686159844057e-05,
|
| 119 |
+
"loss": 0.1953,
|
| 120 |
+
"step": 650
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.0701754385964914,
|
| 124 |
+
"grad_norm": 4.068276882171631,
|
| 125 |
+
"learning_rate": 1.5409356725146202e-05,
|
| 126 |
+
"loss": 0.1248,
|
| 127 |
+
"step": 700
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.2894736842105265,
|
| 131 |
+
"grad_norm": 0.6126830577850342,
|
| 132 |
+
"learning_rate": 1.4922027290448344e-05,
|
| 133 |
+
"loss": 0.1289,
|
| 134 |
+
"step": 750
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.5087719298245617,
|
| 138 |
+
"grad_norm": 6.3761796951293945,
|
| 139 |
+
"learning_rate": 1.4434697855750488e-05,
|
| 140 |
+
"loss": 0.1117,
|
| 141 |
+
"step": 800
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.7280701754385968,
|
| 145 |
+
"grad_norm": 4.546126365661621,
|
| 146 |
+
"learning_rate": 1.3947368421052631e-05,
|
| 147 |
+
"loss": 0.1491,
|
| 148 |
+
"step": 850
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.9473684210526314,
|
| 152 |
+
"grad_norm": 4.006205081939697,
|
| 153 |
+
"learning_rate": 1.3460038986354777e-05,
|
| 154 |
+
"loss": 0.1345,
|
| 155 |
+
"step": 900
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 4.166666666666667,
|
| 159 |
+
"grad_norm": 2.2420225143432617,
|
| 160 |
+
"learning_rate": 1.297270955165692e-05,
|
| 161 |
+
"loss": 0.0904,
|
| 162 |
+
"step": 950
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 4.385964912280702,
|
| 166 |
+
"grad_norm": 3.2088565826416016,
|
| 167 |
+
"learning_rate": 1.2485380116959064e-05,
|
| 168 |
+
"loss": 0.0996,
|
| 169 |
+
"step": 1000
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 4.385964912280702,
|
| 173 |
+
"eval_accuracy": 0.9115612648221344,
|
| 174 |
+
"eval_auc": 0.9578519002979434,
|
| 175 |
+
"eval_f1": 0.8182741116751269,
|
| 176 |
+
"eval_false_negatives": 92,
|
| 177 |
+
"eval_false_positives": 87,
|
| 178 |
+
"eval_loss": 0.3153094947338104,
|
| 179 |
+
"eval_npv": 0.940026075619296,
|
| 180 |
+
"eval_precision": 0.8224489795918367,
|
| 181 |
+
"eval_recall": 0.8141414141414142,
|
| 182 |
+
"eval_runtime": 7.4628,
|
| 183 |
+
"eval_samples_per_second": 271.212,
|
| 184 |
+
"eval_specificity": 0.9431000654022237,
|
| 185 |
+
"eval_steps_per_second": 4.288,
|
| 186 |
+
"eval_true_negatives": 1442,
|
| 187 |
+
"eval_true_positives": 403,
|
| 188 |
+
"step": 1000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 4.605263157894737,
|
| 192 |
+
"grad_norm": 10.743696212768555,
|
| 193 |
+
"learning_rate": 1.1998050682261208e-05,
|
| 194 |
+
"loss": 0.0909,
|
| 195 |
+
"step": 1050
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 4.824561403508772,
|
| 199 |
+
"grad_norm": 1.6505861282348633,
|
| 200 |
+
"learning_rate": 1.1510721247563355e-05,
|
| 201 |
+
"loss": 0.088,
|
| 202 |
+
"step": 1100
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 5.043859649122807,
|
| 206 |
+
"grad_norm": 6.567928314208984,
|
| 207 |
+
"learning_rate": 1.1023391812865499e-05,
|
| 208 |
+
"loss": 0.1129,
|
| 209 |
+
"step": 1150
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 5.2631578947368425,
|
| 213 |
+
"grad_norm": 3.445005416870117,
|
| 214 |
+
"learning_rate": 1.0536062378167643e-05,
|
| 215 |
+
"loss": 0.0605,
|
| 216 |
+
"step": 1200
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 5.482456140350878,
|
| 220 |
+
"grad_norm": 1.5516074895858765,
|
| 221 |
+
"learning_rate": 1.0048732943469786e-05,
|
| 222 |
+
"loss": 0.0642,
|
| 223 |
+
"step": 1250
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 5.701754385964913,
|
| 227 |
+
"grad_norm": 9.6685209274292,
|
| 228 |
+
"learning_rate": 9.56140350877193e-06,
|
| 229 |
+
"loss": 0.0759,
|
| 230 |
+
"step": 1300
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 5.921052631578947,
|
| 234 |
+
"grad_norm": 12.226219177246094,
|
| 235 |
+
"learning_rate": 9.074074074074075e-06,
|
| 236 |
+
"loss": 0.0806,
|
| 237 |
+
"step": 1350
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 6.140350877192983,
|
| 241 |
+
"grad_norm": 1.530657172203064,
|
| 242 |
+
"learning_rate": 8.586744639376219e-06,
|
| 243 |
+
"loss": 0.0636,
|
| 244 |
+
"step": 1400
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 6.359649122807017,
|
| 248 |
+
"grad_norm": 4.697056293487549,
|
| 249 |
+
"learning_rate": 8.099415204678363e-06,
|
| 250 |
+
"loss": 0.0402,
|
| 251 |
+
"step": 1450
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 6.578947368421053,
|
| 255 |
+
"grad_norm": 0.09656322002410889,
|
| 256 |
+
"learning_rate": 7.612085769980507e-06,
|
| 257 |
+
"loss": 0.0629,
|
| 258 |
+
"step": 1500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 6.578947368421053,
|
| 262 |
+
"eval_accuracy": 0.9179841897233202,
|
| 263 |
+
"eval_auc": 0.9618949468524354,
|
| 264 |
+
"eval_f1": 0.8241525423728814,
|
| 265 |
+
"eval_false_negatives": 106,
|
| 266 |
+
"eval_false_positives": 60,
|
| 267 |
+
"eval_loss": 0.38592642545700073,
|
| 268 |
+
"eval_npv": 0.9326984126984127,
|
| 269 |
+
"eval_precision": 0.8663697104677061,
|
| 270 |
+
"eval_recall": 0.7858585858585858,
|
| 271 |
+
"eval_runtime": 7.4828,
|
| 272 |
+
"eval_samples_per_second": 270.486,
|
| 273 |
+
"eval_specificity": 0.960758665794637,
|
| 274 |
+
"eval_steps_per_second": 4.276,
|
| 275 |
+
"eval_true_negatives": 1469,
|
| 276 |
+
"eval_true_positives": 389,
|
| 277 |
+
"step": 1500
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 6.798245614035087,
|
| 281 |
+
"grad_norm": 36.65667724609375,
|
| 282 |
+
"learning_rate": 7.124756335282652e-06,
|
| 283 |
+
"loss": 0.0606,
|
| 284 |
+
"step": 1550
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 7.017543859649122,
|
| 288 |
+
"grad_norm": 2.3685500621795654,
|
| 289 |
+
"learning_rate": 6.637426900584796e-06,
|
| 290 |
+
"loss": 0.0513,
|
| 291 |
+
"step": 1600
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 7.2368421052631575,
|
| 295 |
+
"grad_norm": 2.3388023376464844,
|
| 296 |
+
"learning_rate": 6.15009746588694e-06,
|
| 297 |
+
"loss": 0.0328,
|
| 298 |
+
"step": 1650
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 7.456140350877193,
|
| 302 |
+
"grad_norm": 0.18014875054359436,
|
| 303 |
+
"learning_rate": 5.662768031189084e-06,
|
| 304 |
+
"loss": 0.0526,
|
| 305 |
+
"step": 1700
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 7.675438596491228,
|
| 309 |
+
"grad_norm": 5.354085445404053,
|
| 310 |
+
"learning_rate": 5.175438596491229e-06,
|
| 311 |
+
"loss": 0.033,
|
| 312 |
+
"step": 1750
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 7.894736842105263,
|
| 316 |
+
"grad_norm": 0.14219924807548523,
|
| 317 |
+
"learning_rate": 4.688109161793373e-06,
|
| 318 |
+
"loss": 0.0286,
|
| 319 |
+
"step": 1800
|
| 320 |
+
},
|
| 321 |
+
{
|
| 322 |
+
"epoch": 8.114035087719298,
|
| 323 |
+
"grad_norm": 0.8848229646682739,
|
| 324 |
+
"learning_rate": 4.200779727095517e-06,
|
| 325 |
+
"loss": 0.023,
|
| 326 |
+
"step": 1850
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 8.333333333333334,
|
| 330 |
+
"grad_norm": 0.09083296358585358,
|
| 331 |
+
"learning_rate": 3.713450292397661e-06,
|
| 332 |
+
"loss": 0.0342,
|
| 333 |
+
"step": 1900
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.552631578947368,
|
| 337 |
+
"grad_norm": 5.818343162536621,
|
| 338 |
+
"learning_rate": 3.2261208576998056e-06,
|
| 339 |
+
"loss": 0.0317,
|
| 340 |
+
"step": 1950
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 8.771929824561404,
|
| 344 |
+
"grad_norm": 0.027245348319411278,
|
| 345 |
+
"learning_rate": 2.7387914230019497e-06,
|
| 346 |
+
"loss": 0.0344,
|
| 347 |
+
"step": 2000
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.771929824561404,
|
| 351 |
+
"eval_accuracy": 0.9169960474308301,
|
| 352 |
+
"eval_auc": 0.9587609251441821,
|
| 353 |
+
"eval_f1": 0.8264462809917356,
|
| 354 |
+
"eval_false_negatives": 95,
|
| 355 |
+
"eval_false_positives": 73,
|
| 356 |
+
"eval_loss": 0.44914552569389343,
|
| 357 |
+
"eval_npv": 0.938749194068343,
|
| 358 |
+
"eval_precision": 0.8456659619450317,
|
| 359 |
+
"eval_recall": 0.8080808080808081,
|
| 360 |
+
"eval_runtime": 7.4877,
|
| 361 |
+
"eval_samples_per_second": 270.31,
|
| 362 |
+
"eval_specificity": 0.9522563767168084,
|
| 363 |
+
"eval_steps_per_second": 4.274,
|
| 364 |
+
"eval_true_negatives": 1456,
|
| 365 |
+
"eval_true_positives": 400,
|
| 366 |
+
"step": 2000
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 8.991228070175438,
|
| 370 |
+
"grad_norm": 6.446545124053955,
|
| 371 |
+
"learning_rate": 2.2514619883040934e-06,
|
| 372 |
+
"loss": 0.031,
|
| 373 |
+
"step": 2050
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 9.210526315789474,
|
| 377 |
+
"grad_norm": 3.8243560791015625,
|
| 378 |
+
"learning_rate": 1.7641325536062378e-06,
|
| 379 |
+
"loss": 0.0322,
|
| 380 |
+
"step": 2100
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 9.429824561403509,
|
| 384 |
+
"grad_norm": 0.14315390586853027,
|
| 385 |
+
"learning_rate": 1.2768031189083821e-06,
|
| 386 |
+
"loss": 0.025,
|
| 387 |
+
"step": 2150
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 9.649122807017545,
|
| 391 |
+
"grad_norm": 0.08034974336624146,
|
| 392 |
+
"learning_rate": 7.894736842105263e-07,
|
| 393 |
+
"loss": 0.0143,
|
| 394 |
+
"step": 2200
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 9.868421052631579,
|
| 398 |
+
"grad_norm": 0.1442982703447342,
|
| 399 |
+
"learning_rate": 3.021442495126706e-07,
|
| 400 |
+
"loss": 0.0121,
|
| 401 |
+
"step": 2250
|
| 402 |
+
}
|
| 403 |
+
],
|
| 404 |
+
"logging_steps": 50,
|
| 405 |
+
"max_steps": 2280,
|
| 406 |
+
"num_input_tokens_seen": 0,
|
| 407 |
+
"num_train_epochs": 10,
|
| 408 |
+
"save_steps": 500,
|
| 409 |
+
"stateful_callbacks": {
|
| 410 |
+
"EarlyStoppingCallback": {
|
| 411 |
+
"args": {
|
| 412 |
+
"early_stopping_patience": 3,
|
| 413 |
+
"early_stopping_threshold": 0.001
|
| 414 |
+
},
|
| 415 |
+
"attributes": {
|
| 416 |
+
"early_stopping_patience_counter": 0
|
| 417 |
+
}
|
| 418 |
+
},
|
| 419 |
+
"TrainerControl": {
|
| 420 |
+
"args": {
|
| 421 |
+
"should_epoch_stop": false,
|
| 422 |
+
"should_evaluate": false,
|
| 423 |
+
"should_log": false,
|
| 424 |
+
"should_save": true,
|
| 425 |
+
"should_training_stop": true
|
| 426 |
+
},
|
| 427 |
+
"attributes": {}
|
| 428 |
+
}
|
| 429 |
+
},
|
| 430 |
+
"total_flos": 9579873525657600.0,
|
| 431 |
+
"train_batch_size": 32,
|
| 432 |
+
"trial_name": null,
|
| 433 |
+
"trial_params": null
|
| 434 |
+
}
|
checkpoint-2280/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fa0b6e71cf97dcf125328dcde4d9907006c97167c4f41af8ac98ea5f88ac6b4
|
| 3 |
+
size 5713
|
checkpoint-2280/vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
config.json
ADDED
|
@@ -0,0 +1,26 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"BertForSequenceClassification"
|
| 4 |
+
],
|
| 5 |
+
"attention_probs_dropout_prob": 0.1,
|
| 6 |
+
"classifier_dropout": null,
|
| 7 |
+
"gradient_checkpointing": false,
|
| 8 |
+
"hidden_act": "gelu",
|
| 9 |
+
"hidden_dropout_prob": 0.1,
|
| 10 |
+
"hidden_size": 768,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"intermediate_size": 3072,
|
| 13 |
+
"layer_norm_eps": 1e-12,
|
| 14 |
+
"max_position_embeddings": 512,
|
| 15 |
+
"model_type": "bert",
|
| 16 |
+
"num_attention_heads": 12,
|
| 17 |
+
"num_hidden_layers": 12,
|
| 18 |
+
"pad_token_id": 0,
|
| 19 |
+
"position_embedding_type": "absolute",
|
| 20 |
+
"problem_type": "single_label_classification",
|
| 21 |
+
"torch_dtype": "float32",
|
| 22 |
+
"transformers_version": "4.51.3",
|
| 23 |
+
"type_vocab_size": 2,
|
| 24 |
+
"use_cache": true,
|
| 25 |
+
"vocab_size": 30522
|
| 26 |
+
}
|
eval_metrics.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eval_loss": 0.44914552569389343,
|
| 3 |
+
"eval_accuracy": 0.9169960474308301,
|
| 4 |
+
"eval_precision": 0.8456659619450317,
|
| 5 |
+
"eval_recall": 0.8080808080808081,
|
| 6 |
+
"eval_f1": 0.8264462809917356,
|
| 7 |
+
"eval_auc": 0.9587609251441821,
|
| 8 |
+
"eval_specificity": 0.9522563767168084,
|
| 9 |
+
"eval_npv": 0.938749194068343,
|
| 10 |
+
"eval_true_positives": 400,
|
| 11 |
+
"eval_true_negatives": 1456,
|
| 12 |
+
"eval_false_positives": 73,
|
| 13 |
+
"eval_false_negatives": 95,
|
| 14 |
+
"eval_runtime": 7.4496,
|
| 15 |
+
"eval_samples_per_second": 271.691,
|
| 16 |
+
"eval_steps_per_second": 4.296,
|
| 17 |
+
"epoch": 10.0
|
| 18 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:41590102e903624f664df64252a13ee88e911b1cb5be074c17b4957d02712d72
|
| 3 |
+
size 437958648
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"cls_token": {
|
| 3 |
+
"content": "[CLS]",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"mask_token": {
|
| 10 |
+
"content": "[MASK]",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "[PAD]",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"sep_token": {
|
| 24 |
+
"content": "[SEP]",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": false,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
},
|
| 30 |
+
"unk_token": {
|
| 31 |
+
"content": "[UNK]",
|
| 32 |
+
"lstrip": false,
|
| 33 |
+
"normalized": false,
|
| 34 |
+
"rstrip": false,
|
| 35 |
+
"single_word": false
|
| 36 |
+
}
|
| 37 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,58 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"added_tokens_decoder": {
|
| 3 |
+
"0": {
|
| 4 |
+
"content": "[PAD]",
|
| 5 |
+
"lstrip": false,
|
| 6 |
+
"normalized": false,
|
| 7 |
+
"rstrip": false,
|
| 8 |
+
"single_word": false,
|
| 9 |
+
"special": true
|
| 10 |
+
},
|
| 11 |
+
"100": {
|
| 12 |
+
"content": "[UNK]",
|
| 13 |
+
"lstrip": false,
|
| 14 |
+
"normalized": false,
|
| 15 |
+
"rstrip": false,
|
| 16 |
+
"single_word": false,
|
| 17 |
+
"special": true
|
| 18 |
+
},
|
| 19 |
+
"101": {
|
| 20 |
+
"content": "[CLS]",
|
| 21 |
+
"lstrip": false,
|
| 22 |
+
"normalized": false,
|
| 23 |
+
"rstrip": false,
|
| 24 |
+
"single_word": false,
|
| 25 |
+
"special": true
|
| 26 |
+
},
|
| 27 |
+
"102": {
|
| 28 |
+
"content": "[SEP]",
|
| 29 |
+
"lstrip": false,
|
| 30 |
+
"normalized": false,
|
| 31 |
+
"rstrip": false,
|
| 32 |
+
"single_word": false,
|
| 33 |
+
"special": true
|
| 34 |
+
},
|
| 35 |
+
"103": {
|
| 36 |
+
"content": "[MASK]",
|
| 37 |
+
"lstrip": false,
|
| 38 |
+
"normalized": false,
|
| 39 |
+
"rstrip": false,
|
| 40 |
+
"single_word": false,
|
| 41 |
+
"special": true
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"clean_up_tokenization_spaces": true,
|
| 45 |
+
"cls_token": "[CLS]",
|
| 46 |
+
"do_basic_tokenize": true,
|
| 47 |
+
"do_lower_case": true,
|
| 48 |
+
"extra_special_tokens": {},
|
| 49 |
+
"mask_token": "[MASK]",
|
| 50 |
+
"model_max_length": 512,
|
| 51 |
+
"never_split": null,
|
| 52 |
+
"pad_token": "[PAD]",
|
| 53 |
+
"sep_token": "[SEP]",
|
| 54 |
+
"strip_accents": null,
|
| 55 |
+
"tokenize_chinese_chars": true,
|
| 56 |
+
"tokenizer_class": "BertTokenizer",
|
| 57 |
+
"unk_token": "[UNK]"
|
| 58 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fa0b6e71cf97dcf125328dcde4d9907006c97167c4f41af8ac98ea5f88ac6b4
|
| 3 |
+
size 5713
|
vocab.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|