[DOC-REVIEW] Frames README (#2) #292
Conversation
README.md
Outdated
| <a id="method-create-params-tsdb"></a> | ||
| #### tsdb Backend create Parameters | ||
|
|
||
| - <a id="method-create-tsdb-param-rate"></a>**rate** (Required) — `string` — the ingestion rate TSDB's metric-samples, as `"[0-9]+/[smh]"` (where `s` = seconds, `m` = minutes, and `h` = hours); for example, `"1/s"` (one sample per minute). |
There was a problem hiding this comment.
... the approximate ingestion rate for the TSDB's metric-samples ...
There was a problem hiding this comment.
@talIguaz the next line in this parameter description is "The rate should be calculated according to the slowest expected ingestion rate.". Doesn't this cover what you wanted to add?
There was a problem hiding this comment.
Don't think slowest is necessarily better. i guess an average rate will suite better.
But it depends on if there's a big difference between the expected rates. if so we need to think what will be better, otherwise i guess avg should work.
There was a problem hiding this comment.
@talIguaz currently, in our doc-site TSDB CLI tutorial, Frames v3io/tutorials NB, and Frames README file we say that the TSDB ingestion rate should be set according to the slowest expected ingestion rate. This doc was added based on Golan's guidelines. @dinal please let me know whether we need to change to this guideline, in light of Tal's input, and if so, what do you want the new guideline to be.
…ernal info for schema param (v3io#3) (PR 292 #2d)
|
@talIguaz I finished the second review round and committed the changes to the PR. In addition to changes pertaining to your review comments, I also made additional fixes and improvements — including to add return-value sections for the client constructor and the read method, change the syntax code snippets, and fix and improve the method parameter descriptions (to be continued ...). |
| session : Session | ||
| Session object | ||
| frame_factory : class | ||
| DataFrame factory (currencly pandas and cudf supported) | ||
| DataFrame factory; currently, pandas and cuDF are supported |
There was a problem hiding this comment.
Not sure we want to say that cuDF is officially supported yet
There was a problem hiding this comment.
|
|
||
| All Frames backends that support the `read` method support the following common parameters: | ||
|
|
||
| - <a id="method-read-param-iterator"></a>**iterator** — (Optional) (default: `False`) — `bool` — `True` to return a DataFrames iterator; `False` to return a single DataFrame. |
There was a problem hiding this comment.
better to order the parameters with the optional ones in the end
There was a problem hiding this comment.
@talIguaz I plan to reorder all parameter lists alphabetically and include separate requirement bullets (as I did in the constructor documentation). I just didn't have time to handle this for this PR + I want to do the reordering in a separate commit to simplify diffs.
|
|
||
| - <a id="method-create-param-attrs"></a>**attrs** — A dictionary of `<argument name>: <value>` pairs for passing additional backend-specific parameters (arguments). | ||
|
|
||
| - **Type:** dict |
There was a problem hiding this comment.
The style of the parameter description here is different than the other methods (read,write..). is it on purpose?
There was a problem hiding this comment.
@talIguaz ultimately, all parameter descriptions should have separate type and requirement bullets (and valid and default values bullets where relevant), I just didn't have time to do it for this PR.
No description provided.