# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("ittailup/distilgender-es-2M")
model = AutoModelForSequenceClassification.from_pretrained("ittailup/distilgender-es-2M", device_map="auto")Quick Links
Model Card
Overview
This model card provides details about a trained model, its training process, and evaluation metrics. This information ensures transparency and assists users in understanding the model's performance and behavior.
Training Details
- Training Epochs: The model was trained for 2 epochs.
- Training Steps: The model underwent 1856 training steps.
- Training Runtime: The model's training runtime was approximately 2680.184 seconds.
- Training Speed: The model trained at a rate of 0.692 steps per second and processed approximately 1417.813 samples per second.
- Learning Rate: The learning rate during training was approximately 0.0000095905.
- Training Loss: The average training loss recorded was approximately 0.0184, with a specific loss value of 0.023423514232553285.
Evaluation Details
- Evaluation Loss: The model achieved an evaluation loss of 0.017659155651926994.
- Evaluation Runtime: The evaluation process took approximately 23.8414 seconds.
- Evaluation Speed: The model was evaluated at a rate of 2.055 steps per second, processing approximately 4194.378 samples per second.
Performance Metrics
- Accuracy: The model achieved an accuracy of 0.9951 during evaluation.
- Precision: The precision of the model is approximately 0.9957234121187588.
- Recall: The model's recall is approximately 0.9956533216014078.
- F1-Score: The F1-Score for the model is approximately 0.995688365626595.
- Downloads last month
- 6
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="ittailup/distilgender-es-2M")