|
My Colloquial Telugu Model
|
|
|
|
Overview
|
|
|
|
This model is fine-tuned on colloquial Telugu text to enhance natural language understanding and generation in informal Telugu conversations. It can be used for chatbots, sentiment analysis, text classification, and other NLP tasks.
|
|
|
|
|
|
---
|
|
|
|
Model Details
|
|
|
|
Model Name: My Colloquial Telugu Model
|
|
|
|
Base Model: bert-base-multilingual-cased
|
|
|
|
Training Data: Telugu colloquial dataset containing informal conversations
|
|
|
|
Fine-tuning Details:
|
|
|
|
Epochs: 3
|
|
|
|
Batch Size: 16
|
|
|
|
Optimizer: AdamW
|
|
|
|
|
|
Language: Telugu
|
|
|
|
|
|
|
|
---
|
|
|
|
Usage
|
|
|
|
This model can be used for text generation, classification, and translation.
|
|
|
|
Using the Model in Python
|
|
|
|
To use this model in Python, install the required libraries:
|
|
|
|
pip install transformers torch
|
|
|
|
Then, load the model:
|
|
|
|
from transformers import AutoModel, AutoTokenizer
|
|
|
|
model_name = "your-hf-username/my_colloquial_telugu_model"
|
|
tokenizer = AutoTokenizer.from_pretrained(model_name)
|
|
model = AutoModel.from_pretrained(model_name)
|
|
|
|
text = "ఇవాళ వాతావరణం ఎలా ఉంది?"
|
|
inputs = tokenizer(text, return_tensors="pt")
|
|
outputs = model(**inputs)
|
|
print(outputs)
|
|
|
|
|
|
---
|
|
|
|
Performance
|
|
|
|
Accuracy/F1 Score: TBD (To Be Determined)
|
|
|
|
Limitations:
|
|
|
|
May struggle with highly formal or technical Telugu text.
|
|
|
|
Performance depends on dataset quality and coverage.
|
|
|
|
|
|
|
|
|
|
---
|
|
|
|
License
|
|
|
|
This model is released under the Apache 2.0 license. |