sentence-transformers How to use agraharr/telecom-bge-base-hard-neg with sentence-transformers:
from sentence_transformers import SentenceTransformer
model = SentenceTransformer("agraharr/telecom-bge-base-hard-neg")
sentences = [
"What does DM-RS stand for?",
"In a groupcast scenario, a TX UE restarts its SL inactivity timer for the specific destination L2 ID whenever it receives new data directed to that same destination L2 ID. This mechanism ensures that the allowable transmission time is accurately determined based on the most recent data received. The RX UE, on the other hand, maintains a SL inactivity timer for each destination L2 ID and selects the largest timer value when multiple timers associated with different QoS profiles are configured. This coordination between TX and RX UEs helps maintain effective communication and manage resource allocation efficiently in groupcast transmissions.<|im_end|>",
"For a UE to identify and report the CGI of a known NR target cell while in RRC_CONNECTED state, it must be configured for SA operation mode. The UE shall identify and report the CGI when prompted by the network through the reportCGI command. It will receive one cell indication through *cellForWhichToReportCGI* for this purpose. The UE may utilize autonomous gaps in both downlink and uplink to receive the MIB and SIB1 messages, unless *useAutonomousGaps* is set to false. The identification of the CGI must be completed within a specific time frame, denoted as T~identify_CGI_redcap~, which is derived from the time taken to acquire the MIB and SIB1 messages. Furthermore, during this identification period, the UE must meet certain interruption requirements and ensure that it can detect the necessary signals under specified conditions. This process is crucial for maintaining connectivity and ensuring accurate cell identification in the network.<|im_end|>",
"Demodulation Reference Signal"
]
embeddings = model.encode(sentences)
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [4, 4] 
