File size: 3,581 Bytes
809757b a869d2e 809757b a869d2e 809757b a869d2e 809757b a869d2e 809757b 16573cd 809757b 16573cd 809757b 16573cd 809757b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 |
---
language:
- en
---
# Text Classification GoEmotions
This is a quantized onnx model and is a fined-tuned version of [nreimers/MiniLMv2-L6-H384-distilled-from-BERT-Large](https://huggingface.co/nreimers/MiniLMv2-L6-H384-distilled-from-BERT-Large) on the on the [Jigsaw 1st Kaggle competition](https://www.kaggle.com/competitions/jigsaw-toxic-comment-classification-challenge) dataset using [unitary/toxic-bert](https://huggingface.co/unitary/toxic-bert) as teacher model.
# Load the Model
```py
import os
import numpy as np
import json
from tokenizers import Tokenizer
from onnxruntime import InferenceSession
# !git clone https://huggingface.co/Ngit/MiniLM-L6-toxic-all-labels-onnx
model_name = "Ngit/MiniLM-L6-toxic-all-labels-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 = ["This is pure trash",]
outputs = []
model = InferenceSession("MiniLM-L6-toxic-all-labels-onnx/model_optimized_quantized.onnx", providers=['CUDAExecutionProvider'])
with open(os.path.join("MiniLM-L6-toxic-all-labels-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], dtype=np.int64
),
"attention_mask": np.vstack(
[encoding.attention_mask for encoding in encodings], dtype=np.int64
),
"token_type_ids": np.vstack(
[encoding.type_ids for encoding in encodings], dtype=np.int64
),
}
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: 48
- eval_batch_size: 48
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- warmup_ratio: 0.1
# Metrics (comparison with teacher model)
| Teacher (params) | Student (params) | Set (metric) | Score (teacher) | Score (student) |
|--------------------|-------------|----------|--------| --------|
| unitary/toxic-bert (110M) | MiniLMv2-L6-H384-goemotions-v2-onnx (23M) | Test (ROC_AUC) | 0.98636 | 0.98130 |
# Deployment
Check [this repository](https://github.com/minuva/toxicity-prediction-serverless) to see how to easily deploy this model in a serverless environment with fast CPU inference.
|