Not able to run any python script in Jupiter Notebook Beta environment

Hi Yadav. Thanks for reaching out to Cognite. The query is making individual calls, and looping back over.  It’s. a long ride to 5k. 

Here’s the sequence of operations I would suggest:

1. Fetch from Raw.
2. Group data points by extid
3. Insert the batched data points, and print the output of the payload.

from collections import defaultdict
from cognite.client import CogniteClient
from cognite.client.data_classes.data_modeling import NodeId

client = CogniteClient()

# 1. Fetch RAW rows
all_rows = client.raw.rows.list(
    db_name="IFSDB", 
    table_name="draft-values-test", 
    limit=5000
)

# 2. Group datapoints by externalId in memory to batch them
# Structure: { NodeId: [(timestamp, value), (timestamp, value), ...] }
batched_data = defaultdict(list)

print("Processing and grouping RAW data...")
for row in all_rows:
    sensor = row.columns.get('sensor')
    timestamp = row.columns.get('time_stamp')
    dp_value = row.columns.get('value')

    # Skip row if any critical data is missing
    if not (sensor and timestamp and dp_value is not None):
        continue

    instance_key = NodeId("cdf-fundamentals", f"FirstnameBirthyear:{sensor}")
    batched_data[instance_key].append((timestamp, dp_value))

# 3. Insert the batched datapoints
print("Ingesting datapoints in batches...")
insert_payload = []
for instance_id, dps in batched_data.items():
    insert_payload.append({
        "instance_id": instance_id,
        "datapoints": dps
    })

# Cognite's SDK handles bulk list inserts beautifully
if insert_payload:
    client.time_series.data.insert_multiple(insert_payload)
    print(f"Successfully ingested data for {len(insert_payload)} unique time series.")
else:
    print("No valid datapoints found to insert.")

Let me know if this addresses the issues you’re facing.

Regards,

MB