Comments (7)
from textbrewer.
Thank you for the quick reply! It solves my question.
from textbrewer.
Sorry to bother again, but I have a little problem in reproducing the NER results in the paper.
By using the example script (./examples/conll2003_example/), I can successfully reproduce BERT's result in the supervised learning setting (I got 91.3 vs 91.1 in the paper). But I only get much worse distilled results.
I distill the logits (ce loss) as well as the intermediate hidden states (mse and mmd loss) as described in the paper.
I use the following hyper-parameters: lr=1e-4, batch size=32, warmup steps=0.1, epochs=100.
And I got: F1(of T3) = 84.6, F1(of T6) = 89.7.
Here is the train script:
export OUTPUT_DIR=outputs-model-distill
export BATCH_SIZE=32
export GRAD_ACCUM_STEPS=1
export NUM_EPOCHS=100
export SAVE_STEPS=750
export SEED=42
export MAX_LENGTH=128
export BERT_MODEL_TEACHER=outputs-model-base
export BERT_MODEL_STUDENT=bert-base-cased
mkdir -p $OUTPUT_DIR
python run_ner_distill.py \
--data_dir data \
--model_type bert \
--model_name_or_path $BERT_MODEL_TEACHER \
--model_name_or_path_student $BERT_MODEL_STUDENT \
--output_dir $OUTPUT_DIR \
--max_seq_length $MAX_LENGTH \
--num_train_epochs $NUM_EPOCHS \
--per_gpu_train_batch_size $BATCH_SIZE \
--gradient_accumulation_steps $GRAD_ACCUM_STEPS \
--num_hidden_layers 3 \
--save_steps $SAVE_STEPS \
--learning_rate 1e-4 \
--warmup_steps 0.1 \
--seed $SEED \
--do_distill \
--do_train \
--do_eval \
--do_predict \
--overwrite_output_dir \
--overwrite_cache
I think it should be due to my inappropriate experimental setting. Do you have any idea on that? @airaria
from textbrewer.
from textbrewer.
Hi @airaria, do you have any findings? I tried removing the mmd loss, but it did not help improve the F1; I also tried a smaller learning rate, but it did not help either.
When I used data augmentation on T3 and T6, their F1s could successfully exceed the reported scores. However, as stated in this document, only T4-tiny and T12-nano use data augmentation, so there should be other factors that affect my results orz..
from textbrewer.
from textbrewer.
Thanks a lot! That's very useful!
from textbrewer.
Related Issues (20)
- pre-trained student weights HOT 3
- Where to find gs4210.pkl file or how to generate it ? thanks HOT 2
- interpreting intermediate matches HOT 5
- Show the progress bar when training. HOT 3
- Picking right layers HOT 3
- How about the distillation effect of gpt2 ? HOT 2
- Does it support translation model? HOT 2
- 在VisionTransformer HOT 7
- 关于ner数据的处理 HOT 2
- notebook_examples/msra_ner.ipynb 运行报错 HOT 12
- 不同维度蒸馏有对应的例子吗,从768降到256 HOT 4
- msra_ner.ipynb最后的trainer.evaluate()显示CUDA out of memory,请问训练的显存要求是多大?十分感谢! HOT 2
- 老师,您好,请问有多任务多教师的蒸馏的demo吗? HOT 4
- 老师您好,我想问一下,比如roberta蒸馏到tinybert,中间的hidden是通过线性层拉到同样的维度去算mse,那在推理的时候岂不是这些经过梯度更新的线性层毫无作用?那请问这些线性层仅仅就是为了调整维度? HOT 2
- 蒸馏后的模型进行evaluate,报错AxisError: axis 2 is out of bounds for array of dimension 1 HOT 5
- 可以使用chatgpt蒸馏到bert或者T5吗? HOT 2
- 麻烦问下,目前支持llama模型吗 HOT 2
- 请问支持BERT-of-Theseus的蒸馏方式吗 HOT 3
- 学生模型权重初始化问题 HOT 2
- TextBrewer/src/textbrewer/distiller_utils.py get_outputs_from_batch fails tocheck dicts properly for maskedLM HOT 4
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from textbrewer.