Product Engineering Review: Recommendations_product_search_OpenAI_Python

[chetan-hirapara]

Reviewer 1 comments:

• sqle queries are directly executed.
• Plotting use plotty.

Reviewer 1 suggestions:

• tdml supports all fundamental plots, Which can be used for plotting Scatter and bar graph.
• VectorDistane function is supported in teradataml.
• Lot of data movements
  are present, If created
  table are not required
  in future, then volatile
  feature can be used.

Reviewer 2 comments:

• SQL Queries are used.

Reviewer 2 suggestions:

• teradataml supports groupby and join.Create two dataframes and do a join instead of doing a SQL Query
• One of the cell explains about products for snacks but uses dataframe.sort . Instead use dataframe.drop_duplicate which shows all
  the distinct products.

Reviewer 3 comments:

• In section 6, TD_VECTORDISTANCE is run using SQL query.
• In section 7.1, scatter plot is plotted using plotly.express
• In section7.1, pandas merge is used to merge dfs
• Pandas read_csv is used at multiple places

Reviewer 3 suggestions:

• In section 6, pythonic function VectorDistance can be used.
• In section 7.1, tdml scatter plots can be used
• In section 7.1, merging of dataframes can be done by using tdml DF’s merge
• read_csv of tdml can be used wherever possible.
• read_csv of tdml can be used wherever possible.

[chetan-hirapara]

Reviewer 1 comments:

• sqle queries are directly executed. --- Fixed in pass6
• Plotting use plotty. --- Scatterplot with bubble and hover details are note supported in teradataml hence used plotly functions. Same for heatmap._

Reviewer 1 suggestions:

• tdml supports all fundamental plots, Which can be used for plotting Scatter and bar graph. --- _Scatterplot with bubble and hover details are note supported in teradataml hence used plotly functions.

• VectorDistane function is supported in teradataml. --- Fixed in pass6

• Lot of data movements are present, If created table are not required in future, then volatile feature can be used. --- Thanks, fixed

Reviewer 2 comments:

• SQL Queries are used. --- Fixed in pass6

Reviewer 2 suggestions:

• teradataml supports groupby and join. Create two dataframes and do a join instead of doing a SQL Query --- Fixed in pass6
• One of the cell explains about products for snacks but uses dataframe.sort . Instead use dataframe.drop_duplicate which shows all
  the distinct products. --- There was not a duplicate products, sort is used just for display proeucts in proper sequential IDs, instead of random numbers.

Reviewer 3 comments:

• In section 6, TD_VECTORDISTANCE is run using SQL query. --- Fixed in pass6
• In section 7.1, scatter plot is plotted using plotly.express --- Scatterplot with bubble and hover details are note supported in teradataml hence used plotly functions. Same for heatmap.
• In section7.1, pandas merge is used to merge dfs --- As this dataframes has more than 3000 columns, which is not able to handle by teradataml dataframe. teradataml dataframe is throwing error: OperationalError: [Version 17.20.0.0] [Session 1281] [Teradata Database] [Error 3919] Table has too many columns.
• Pandas read_csv is used at multiple places --- This is for batch/chunking approach. we have to generate the embeddings in batches

Reviewer 3 suggestions:

• In section 6, pythonic function VectorDistance can be used. --- We have VectorDistance in teradataml so no need to use python one
• In section 7.1, tdml scatter plots can be used --- Scatterplot with bubble and hover details are note supported in teradataml hence used plotly functions. Same for heatmap.
• In section 7.1, merging of dataframes can be done by using tdml DF’s merge --- As this dataframes has more than 3000 columns, which is not able to handle by teradataml dataframe. teradataml dataframe is throwing error: OperationalError: [Version 17.20.0.0] [Session 1281] [Teradata Database] [Error 3919] Table has too many columns.
• read_csv of tdml can be used wherever possible. --- This is for batch/chunking approach. we have to generate the embeddings in batches