关于自己训练PaddleOCR识别模型后，模型的泛化能力下降的问题求助

[wsybb252237]

自己在进行手写字的训练的时候，使用的是自己数据集（大概4256张图片），使用的预训练模型是“ch_PP-OCRv4_rec_server_infer”，但是训练完以后，在自己的数据集上识别效果不错，但是在其他的汉字的识别上效果较差，这些字是预训练模型应该能够识别出来的字，我想请教一下究竟是什么原因导致的，难道自己的数据集微调后，模型在其他数据上的泛化能力就会变差吗，希望能够得到回复，万分感谢！下面是我的参数：
Global:
debug: false
use_gpu: true
epoch_num: 100
log_smooth_window: 20
print_batch_step: 10
save_model_dir: ./output/rec_ppocr_v4_hgnet_server
save_epoch_step: 10
eval_batch_step: [0, 200]
cal_metric_during_train: true
pretrained_model: /PaddleOCR_server/pretrained/ch_PP-OCRv4_rec_server_train/best_accuracy.pdparams
checkpoints:
save_inference_dir:
use_visualdl: false
infer_img: doc/imgs_words/ch/word_1.jpg
character_dict_path: ppocr/utils/ppocr_keys_v1.txt
max_text_length: &max_text_length 25
infer_mode: false
use_space_char: true
distributed: true
save_res_path: ./output/rec/predicts_ppocrv3.txt
d2s_train_image_shape: [3, 48, 320]

Optimizer:
name: Adam
beta1: 0.9
beta2: 0.999
lr:
name: Piecewise

learning_rate: [0.00002, 0.000005]

decay_epochs: [ 70 ]           # 在第30和45个epoch进行学习率衰减
values: [ 0.00002, 0.000005 ]      # 对应两个阶段的学习率
warmup_epoch: 5

regularizer:
name: L2
factor: 3.0e-05

Architecture:
model_type: rec
algorithm: SVTR_HGNet
Transform:
Backbone:
name: PPHGNet_small
Head:
name: MultiHead
head_list:
- CTCHead:
Neck:
name: svtr
dims: 120
depth: 2
hidden_dims: 120
kernel_size: [1, 3]
use_guide: True
Head:
fc_decay: 0.00001
- NRTRHead:
nrtr_dim: 384
max_text_length: *max_text_length

Loss:
name: MultiLoss
loss_config_list:
- CTCLoss:
- NRTRLoss:

PostProcess:
name: CTCLabelDecode

Metric:
name: RecMetric
main_indicator: acc

Train:
dataset:
name: MultiScaleDataSet
ds_width: false
data_dir: ./train_data/
ext_op_transform_idx: 1
label_file_list:
- ./train_data/rec/train.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false

-去除 RecConAug 增广

- RecConAug:

prob: 0.5

ext_data_num: 2

image_shape: [48, 320, 3]

max_text_length: *max_text_length

- RecAug:
- MultiLabelEncode:
    gtc_encode: NRTRLabelEncode
- KeepKeys:
    keep_keys:
    - image
    - label_ctc
    - label_gtc
    - length
    - valid_ratio

sampler:
name: MultiScaleSampler
scales: [[320, 32], [320, 48], [320, 64]]
first_bs: &bs 64
fix_bs: false
divided_factor: [8, 16] # w, h
is_training: True
loader:
shuffle: true
batch_size_per_card: 8
drop_last: true
num_workers: 8
Eval:
dataset:
name: SimpleDataSet
data_dir: ./train_data
label_file_list:
- ./train_data/rec/val.txt
transforms:
- DecodeImage:
img_mode: BGR
channel_first: false
- MultiLabelEncode:
gtc_encode: NRTRLabelEncode
- RecResizeImg:
image_shape: [3, 48, 320]
- KeepKeys:
keep_keys:
- image
- label_ctc
- label_gtc
- length
- valid_ratio
loader:
shuffle: false
drop_last: false
batch_size_per_card: 8
num_workers: 4