Upload README.md

5e3ec76 almost 2 years ago

No virus

4.15 kB

	---
	language:
	- en
	thumbnail: https://avatars3.githubusercontent.com/u/32437151?s=460&u=4ec59abc8d21d5feea3dab323d23a5860e6996a4&v=4
	tags:
	- text-classification
	- emotion
	- pytorch
	license: apache-2.0
	datasets:
	- emotion
	metrics:
	- Accuracy, F1 Score
	model-index:
	- name: bhadresh-savani/distilbert-base-uncased-emotion
	results:
	- task:
	type: text-classification
	name: Text Classification
	dataset:
	name: emotion
	type: emotion
	config: default
	split: test
	metrics:
	- name: Accuracy
	type: accuracy
	value: 0.927
	verified: true
	- name: Precision Macro
	type: precision
	value: 0.8880230732280744
	verified: true
	- name: Precision Micro
	type: precision
	value: 0.927
	verified: true
	- name: Precision Weighted
	type: precision
	value: 0.9272902840835793
	verified: true
	- name: Recall Macro
	type: recall
	value: 0.8790126653780703
	verified: true
	- name: Recall Micro
	type: recall
	value: 0.927
	verified: true
	- name: Recall Weighted
	type: recall
	value: 0.927
	verified: true
	- name: F1 Macro
	type: f1
	value: 0.8825061528287809
	verified: true
	- name: F1 Micro
	type: f1
	value: 0.927
	verified: true
	- name: F1 Weighted
	type: f1
	value: 0.926876082854655
	verified: true
	- name: loss
	type: loss
	value: 0.17403268814086914
	verified: true
	---
	# Distilbert-base-uncased-emotion

	## Model description:
	[Distilbert](https://arxiv.org/abs/1910.01108) is created with knowledge distillation during the pre-training phase which reduces the size of a BERT model by 40%, while retaining 97% of its language understanding. It's smaller, faster than Bert and any other Bert-based model.

	[Distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) finetuned on the emotion dataset using HuggingFace Trainer with below Hyperparameters
	```
	learning rate 2e-5,
	batch size 64,
	num_train_epochs=8,
	```

	## Model Performance Comparision on Emotion Dataset from Twitter:

	\| Model \| Accuracy \| F1 Score \| Test Sample per Second \|
	\| --- \| --- \| --- \| --- \|
	\| [Distilbert-base-uncased-emotion](https://huggingface.co/bhadresh-savani/distilbert-base-uncased-emotion) \| 93.8 \| 93.79 \| 398.69 \|
	\| [Bert-base-uncased-emotion](https://huggingface.co/bhadresh-savani/bert-base-uncased-emotion) \| 94.05 \| 94.06 \| 190.152 \|
	\| [Roberta-base-emotion](https://huggingface.co/bhadresh-savani/roberta-base-emotion) \| 93.95 \| 93.97\| 195.639 \|
	\| [Albert-base-v2-emotion](https://huggingface.co/bhadresh-savani/albert-base-v2-emotion) \| 93.6 \| 93.65 \| 182.794 \|

	## How to Use the model:
	```python
	from transformers import pipeline
	classifier = pipeline("text-classification",model='bhadresh-savani/distilbert-base-uncased-emotion', return_all_scores=True)
	prediction = classifier("I love using transformers. The best part is wide range of support and its easy to use", )
	print(prediction)

	"""
	Output:
	[[
	{'label': 'sadness', 'score': 0.0006792712374590337},
	{'label': 'joy', 'score': 0.9959300756454468},
	{'label': 'love', 'score': 0.0009452480007894337},
	{'label': 'anger', 'score': 0.0018055217806249857},
	{'label': 'fear', 'score': 0.00041110432357527316},
	{'label': 'surprise', 'score': 0.0002288572577526793}
	]]
	"""
	```

	## Dataset:
	[Twitter-Sentiment-Analysis](https://huggingface.co/nlp/viewer/?dataset=emotion).

	## Training procedure
	[Colab Notebook](https://github.com/bhadreshpsavani/ExploringSentimentalAnalysis/blob/main/SentimentalAnalysisWithDistilbert.ipynb)

	## Eval results
	```json
	{
	'test_accuracy': 0.938,
	'test_f1': 0.937932884041714,
	'test_loss': 0.1472451239824295,
	'test_mem_cpu_alloc_delta': 0,
	'test_mem_cpu_peaked_delta': 0,
	'test_mem_gpu_alloc_delta': 0,
	'test_mem_gpu_peaked_delta': 163454464,
	'test_runtime': 5.0164,
	'test_samples_per_second': 398.69
	}
	```

	## Reference:
	* [Natural Language Processing with Transformer By Lewis Tunstall, Leandro von Werra, Thomas Wolf](https://learning.oreilly.com/library/view/natural-language-processing/9781098103231/)