Ashwin Prasad ashwinprasadme

Tested Docker commands for deploying Hugging Face LLMs with vLLM on NVIDIA H100 GPUs.

meta-llama/Llama-3.3-70B-Instruct

docker run --runtime nvidia --gpus all \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HUGGING_FACE_HUB_TOKEN=<token>" \
    -p 8002:8000 \
    --ipc=host \

	## finding all columns that have nan:

	droping_list_all=[]
	for j in range(0,7):
	if not df.iloc[:, j].notnull().all():
	droping_list_all.append(j)
	#print(df.iloc[:,j].unique())
	droping_list_all

	######Prediction###############


	lstm.eval()
	train_predict = lstm(dataX.to(device))
	data_predict = train_predict.cpu().data.numpy()
	dataY_plot = dataY.data.numpy()

	## Inverse Normalize
	data_predict = scaler.inverse_transform(data_predict)

	#Create the testing data set
	#Create a new array containing scaled values from index 1543 to 2002
	test_data = scaled_data[training_data_len - 60: , :]
	#Create the data sets x_test and y_test
	x_test = []
	y_test = dataset[training_data_len:, :]
	for i in range(60, len(test_data)):
	x_test.append(test_data[i-60:i, 0])

	# Convert the data to a numpy array

	def normalize_data(df):
	min_max_scaler = sklearn.preprocessing.MinMaxScaler()
	df['open'] = min_max_scaler.fit_transform(df.open.values.reshape(-1,1))
	df['high'] = min_max_scaler.fit_transform(df.high.values.reshape(-1,1))
	df['low'] = min_max_scaler.fit_transform(df.low.values.reshape(-1,1))
	df['close'] = min_max_scaler.fit_transform(df['close'].values.reshape(-1,1))
	return df

	import numpy as np
	import matplotlib.pyplot as plt
	plt.style.use('fivethirtyeight')
	import pandas as pd
	from sklearn.preprocessing import MinMaxScaler
	from keras.models import Sequential
	from keras.layers import Dense, LSTM, Dropout, GRU, Bidirectional
	from keras.optimizers import SGD
	import math
	from sklearn.metrics import mean_squared_error

	df = df[df.symbol == 'GOOG']
	df.drop(['symbol'],1,inplace=True)
	df.head()