Dec 6, 2022
3 min read
Open AI Chat with new GPT-3.5
Pre-requisites
- Open AI Account
- http://chat.openai.com
- Get a developer key
What is GPT 3.5?
Models referred to as "GPT 3.5"
GPT-3.5 series is a series of models that was trained on a blend of text and code from before Q4 2021. The following models are in the GPT-3.5 series:code-davinci-002 is a base model, so good for pure code-completion tasks
text-davinci-002 is an InstructGPT model based on code-davinci-002
text-davinci-003 is an improvement on text-davinci-002- Information available at https://beta.openai.com/docs/model-index-for-researchers
- Look at the limitation and details about chat in this page — https://openai.com/blog/chatgpt/
- Here we are going to look into writing ML code using GPT-3.5
Create code by asking question
- Create a classification model for titanic dataset with Azure Machine learning Automated ML
# Import the necessary libraries
from azureml.core import Workspace, Dataset
from azureml.train.automl import AutoMLConfig# Load the titanic data into a Dataset object
dataset = Dataset.get_by_name(workspace=workspace, name="titanic")# Split the data into training and testing sets
train, test = dataset.random_split(percentage=0.8, seed=1)# Define the AutoML configuration
automl_config = AutoMLConfig(task="classification",
primary_metric="AUC_weighted",
max_time_sec=3600,
n_cross_validations=5,
debug_log="automl_errors.log",
verbosity=logging.INFO,
X=train.drop_columns(columns=["survived"]),
y=train["survived"],
X_valid=test.drop_columns(columns=["survived"]),
y_valid=test["survived"])# Train the model using the AutoML configuration
automl_run = AutoMLRun(workspace=workspace, config=automl_config)
automl_run.submit(show_output=True)
automl_run.wait_for_completion(show_output=True)# Get the best model from the AutoML run
best_model = automl_run.get_output()# Evaluate the performance of the best model
perf = best_model.predict(test.drop_columns(columns=["survived"]))
print(perf.auc())
- Now next to try Deep learning models
- So here is my question
# Import the necessary libraries
import pandas as pd
from keras.models import Sequential
from keras.layers import Dense# Load the titanic data into a Pandas DataFrame
df = pd.read_csv("titanic.csv")# Split the data into training and testing sets
train, test = df.split_frame(ratios=[0.8])# Define the model architecture
model = Sequential()
model.add(Dense(32, input_dim=6, activation="relu"))
model.add(Dense(16, activation="relu"))
model.add(Dense(1, activation="sigmoid"))# Compile the model
model.compile(loss="binary_crossentropy", optimizer="adam", metrics=["accuracy"])# Train the model
model.fit(x=train.drop("survived", axis=1), y=train["survived"], epochs=10, batch_size=32)# Evaluate the performance of the model
_, acc = model.evaluate(x=test.drop("survived", axis=1), y=test["survived"])
print("Test accuracy:", acc)
- Here is another open source model called yolov5
- question — Can you create me a yolov5 deep learning code?
# Import the necessary libraries
import torch
from torch import nn
import torchvision
from torchvision import models# Define the model architecture
model = models.yolo5(num_classes=80)# Set the model to train mode
model.train()# Define the loss function
criterion = nn.BCEWithLogitsLoss()# Use Adam as the optimizer
optimizer = torch.optim.Adam(model.parameters())# Train the model
for i in range(num_epochs):
for images, labels in train_dataloader:
# Forward pass
outputs = model(images) # Calculate the loss
loss = criterion(outputs, labels) # Backward pass
optimizer.zero_grad()
loss.backward()
optimizer.step()# Set the model to evaluation mode
model.eval()# Evaluate the performance of the model
for images, labels in test_dataloader:
# Forward pass
outputs = model(images) # Calculate the loss
loss = criterion(outputs, labels) # Print the loss
print("Loss:", loss.item())
- This is very interesting, and I am sure this will be very useful for developers and data scientists to create code by asking questions
