metadata

datasets:
  - go_emotions
language:
  - en
library_name: transformers
model-index:
  - name: text-classification-goemotions
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: go_emotions
          type: multilabel_classification
          config: simplified
          split: test
          args: simplified
        metrics:
          - name: F1
            type: f1
            value: 0.482

Text Classification GoEmotions

This a ONNX quantized model and is fined-tuned version of nreimers/MiniLMv2-L6-H384-distilled-from-RoBERTa-Large on the on the go_emotions dataset using tasinho/text-classification-goemotions as teacher model.

Usage

Transformers

No-transformers

Installation

pip install tokenizers
pip install onnxruntime
git clone https://huggingface.co/minuva/MiniLMv2-goemotions-v2-onnx

Load the Model

import os
import numpy as np
import json

from tokenizers import Tokenizer
from onnxruntime import InferenceSession


model_name = "minuva/MiniLMv2-goemotions-v2-onnx"

tokenizer = Tokenizer.from_pretrained(model_name)
tokenizer.enable_padding(
    pad_token="<pad>",
    pad_id=1,
)
tokenizer.enable_truncation(max_length=256)
batch_size = 16

texts = ["I am angry",]
outputs = []
model = InferenceSession("MiniLMv2-goemotions-v2-onnx/model_optimized_quantized.onnx", providers=['CUDAExecutionProvider'])

with open(os.path.join("MiniLMv2-goemotions-v2-onnx", "config.json"), "r") as f:
            config = json.load(f)

output_names = [output.name for output in model.get_outputs()]
input_names = [input.name for input in model.get_inputs()]

for subtexts in np.array_split(np.array(texts), len(texts) // batch_size + 1):
            encodings = tokenizer.encode_batch(list(subtexts))
            inputs = {
                "input_ids": np.vstack(
                    [encoding.ids for encoding in encodings],
                ),
                "attention_mask": np.vstack(
                    [encoding.attention_mask for encoding in encodings],
                ),
                "token_type_ids": np.vstack(
                    [encoding.type_ids for encoding in encodings],
                ),
            }

            for input_name in input_names:
                if input_name not in inputs:
                    raise ValueError(f"Input name {input_name} not found in inputs")

            inputs = {input_name: inputs[input_name] for input_name in input_names}
            output = np.squeeze(
                np.stack(
                    model.run(output_names=output_names, input_feed=inputs)
                ),
                axis=0,
            )
            outputs.append(output)

outputs = np.concatenate(outputs, axis=0)
scores = 1 / (1 + np.exp(-outputs))
results = []
for item in scores:
    labels = []
    scores = []
    for idx, s in enumerate(item):
        labels.append(config["id2label"][str(idx)])
        scores.append(float(s))
    results.append({"labels": labels, "scores": scores})

results

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 6e-05
train_batch_size: 64
eval_batch_size: 64
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear

Metrics (comparison with teacher model)

Teacher (params)	Student (params)	Set	Score (teacher)	Score (student)
tasinhoque/text-classification-goemotions (355M)	MiniLMv2-L6-H384-goemotions-v2-onnx	Validation	0.514252	0.4780
tasinhoque/text-classification-goemotions (33M)	MiniLMv2-L6-H384-goemotions-v2-onnx (original model)	Test	0.501937	0.482

Deployment

Check our repository to see how to easily deploy this model in a serverless environment with fast CPU inference and light resource utilization.