Datasets:

asoria

/

dataset-viewer-github-issues

like 0

feat: 🎸 allow empty features

Anyway: we don't check if the number of features correspond to the number of columns in the rows, so... why not let the features be empty. The client will have to guess the types.

feat: 🎸 allow empty features: Anyway: we don't check if the number of features correspond to the number of columns in the rows, so... why not let the features be empty. The client will have to guess the types.

closed

2021-10-12T14:01:46Z

2021-10-12T14:59:38Z

2021-10-12T14:59:35Z

severo

feat: 🎸 add /webhook endpoint

See https://github.com/huggingface/datasets-preview-backend/issues/36

feat: 🎸 add /webhook endpoint: See https://github.com/huggingface/datasets-preview-backend/issues/36

closed

2021-10-12T09:58:09Z

2021-10-12T11:24:27Z

2021-10-12T11:17:50Z

severo

feat: 🎸 support allenai/c4 dataset

See https://github.com/huggingface/datasets/issues/2859 and https://github.com/huggingface/datasets-preview-backend/issues/17

feat: 🎸 support allenai/c4 dataset: See https://github.com/huggingface/datasets/issues/2859 and https://github.com/huggingface/datasets-preview-backend/issues/17

closed

2021-10-12T08:41:03Z

2021-10-12T08:41:42Z

2021-10-12T08:41:41Z

severo

Support image datasets

Some examples we want to support - Array2D - [x] `mnist` - https://datasets-preview.huggingface.tech/rows?dataset=mnist - Array3D - [x] `cifar10` - https://datasets-preview.huggingface.tech/rows?dataset=cifar10 - [x] `cifar100` - https://datasets-preview.huggingface.tech/rows?dataset=cifar100 - local files: - [x] `food101` - https://datasets-preview.huggingface.tech/rows?dataset=food101 - remote files (URL): - [ ] `compguesswhat` - [x] `severo/wit` - https://datasets-preview.huggingface.tech/rows?dataset=severo/wit

Support image datasets: Some examples we want to support - Array2D - [x] `mnist` - https://datasets-preview.huggingface.tech/rows?dataset=mnist - Array3D - [x] `cifar10` - https://datasets-preview.huggingface.tech/rows?dataset=cifar10 - [x] `cifar100` - https://datasets-preview.huggingface.tech/rows?dataset=cifar100 - local files: - [x] `food101` - https://datasets-preview.huggingface.tech/rows?dataset=food101 - remote files (URL): - [ ] `compguesswhat` - [x] `severo/wit` - https://datasets-preview.huggingface.tech/rows?dataset=severo/wit

closed

2021-10-12T08:20:12Z

2021-10-21T15:30:22Z

2021-10-21T15:30:22Z

severo

feat: 🎸 only cache one entry per dataset

Before: every endpoint call generated a cache entry. Now: all the endpoints calls related to a dataset use the same cached value (which takes longer to compute). The benefits are a simpler code, and most importantly: it's easier to manage cache consistency (everything is OK for a dataset, or nothing) BREAKING CHANGE: 🧨 the /cache, /cache-reports and /valid responses have changed.

feat: 🎸 only cache one entry per dataset: Before: every endpoint call generated a cache entry. Now: all the endpoints calls related to a dataset use the same cached value (which takes longer to compute). The benefits are a simpler code, and most importantly: it's easier to manage cache consistency (everything is OK for a dataset, or nothing) BREAKING CHANGE: 🧨 the /cache, /cache-reports and /valid responses have changed.

closed

2021-10-11T16:20:09Z

2021-10-11T16:23:24Z

2021-10-11T16:23:23Z

severo

Refactor

Refactor:

closed

2021-10-08T17:58:27Z

2021-10-08T17:59:05Z

2021-10-08T17:59:04Z

severo

feat: 🎸 remove benchmark related code

It's covered by the /cache-reports endpoint

feat: 🎸 remove benchmark related code: It's covered by the /cache-reports endpoint

closed

2021-10-08T08:43:52Z

2021-10-08T08:44:00Z

2021-10-08T08:43:59Z

severo

No cache expiration

No cache expiration:

closed

2021-10-08T08:10:41Z

2021-10-08T08:35:30Z

2021-10-08T08:35:29Z

severo

Add valid endpoint

Fixes #24

Add valid endpoint: Fixes #24

closed

2021-10-04T19:51:26Z

2021-10-04T19:51:56Z

2021-10-04T19:51:55Z

severo

Cache functions and responses

Cache functions and responses:

closed

2021-10-04T16:23:56Z

2021-10-04T16:24:46Z

2021-10-04T16:24:45Z

severo

raft - 404

https://datasets-preview.huggingface.tech/rows?dataset=raft&config=ade_corpus_v2&split=test

raft - 404: https://datasets-preview.huggingface.tech/rows?dataset=raft&config=ade_corpus_v2&split=test

closed

2021-10-04T12:33:03Z

2021-10-12T08:52:15Z

2021-10-12T08:52:15Z

severo

Should the features be associated to a split, instead of a config?

For now, we assume that all the splits of a config will share the same features, but it seems that it's not necessarily the case (https://github.com/huggingface/datasets/issues/2968). Am I right @lhoestq ? Is there any example of such a dataset on the hub or in the canonical ones?

Should the features be associated to a split, instead of a config?: For now, we assume that all the splits of a config will share the same features, but it seems that it's not necessarily the case (https://github.com/huggingface/datasets/issues/2968). Am I right @lhoestq ? Is there any example of such a dataset on the hub or in the canonical ones?

closed

2021-10-01T18:14:53Z

2021-10-05T09:25:04Z

2021-10-05T09:25:04Z

severo

/splits does not error when no config exists and a wrong config is passed

https://datasets-preview.huggingface.tech/splits?dataset=sent_comp&config=doesnotexist returns: ``` { "splits": [ { "dataset": "sent_comp", "config": "doesnotexist", "split": "validation" }, { "dataset": "sent_comp", "config": "doesnotexist", "split": "train" } ] } ``` instead of giving an error. As https://datasets-preview.huggingface.tech/configs?dataset=sent_comp returns ``` { "configs": [ { "dataset": "sent_comp", "config": "default" } ] } ``` the only allowed `config` parameter should be `default`.

/splits does not error when no config exists and a wrong config is passed: https://datasets-preview.huggingface.tech/splits?dataset=sent_comp&config=doesnotexist returns: ``` { "splits": [ { "dataset": "sent_comp", "config": "doesnotexist", "split": "validation" }, { "dataset": "sent_comp", "config": "doesnotexist", "split": "train" } ] } ``` instead of giving an error. As https://datasets-preview.huggingface.tech/configs?dataset=sent_comp returns ``` { "configs": [ { "dataset": "sent_comp", "config": "default" } ] } ``` the only allowed `config` parameter should be `default`.

closed

2021-10-01T09:52:40Z

2022-09-16T20:09:54Z

2022-09-16T20:09:53Z

severo

generate infos

fixes #52

generate infos: fixes #52

closed

2021-09-27T13:17:04Z

2021-09-27T13:21:01Z

2021-09-27T13:21:00Z

severo

Regenerate dataset-info instead of loading it?

Currently, getting the rows with `/rows` requires a previous (internal) call to `/infos` to get the features (type of the columns). But sometimes the dataset-info.json file is missing, or not coherent with the dataset script (for example: https://huggingface.co/datasets/lhoestq/custom_squad/tree/main), while we are using `datasets.get_dataset_infos()`, which only loads the exported dataset-info.json files: https://github.com/huggingface/datasets-preview-backend/blob/c2a78e7ce8e36cdf579fea805535fa9ef84a2027/src/datasets_preview_backend/queries/infos.py#L45 https://github.com/huggingface/datasets/blob/26ff41aa3a642e46489db9e95be1e9a8c4e64bea/src/datasets/inspect.py#L115 We might want to call `._info()` from the builder to get the info, and features, instead of relying on the dataset-info.json file.

Regenerate dataset-info instead of loading it?: Currently, getting the rows with `/rows` requires a previous (internal) call to `/infos` to get the features (type of the columns). But sometimes the dataset-info.json file is missing, or not coherent with the dataset script (for example: https://huggingface.co/datasets/lhoestq/custom_squad/tree/main), while we are using `datasets.get_dataset_infos()`, which only loads the exported dataset-info.json files: https://github.com/huggingface/datasets-preview-backend/blob/c2a78e7ce8e36cdf579fea805535fa9ef84a2027/src/datasets_preview_backend/queries/infos.py#L45 https://github.com/huggingface/datasets/blob/26ff41aa3a642e46489db9e95be1e9a8c4e64bea/src/datasets/inspect.py#L115 We might want to call `._info()` from the builder to get the info, and features, instead of relying on the dataset-info.json file.

closed

2021-09-27T11:28:13Z

2021-09-27T13:21:00Z

2021-09-27T13:21:00Z

severo

cache both the functions returns and the endpoints results

Currently only the endpoints results are cached. We use them inside the code to get quick results by taking advantage of the cache, but it's not their aim, and we have to parse / decode. It would be better to directly cache the results of the functions (memoïze). Also: we could cache the raised exceptions as here: https://github.com/peterbe/django-cache-memoize/blob/4da1ba4639774426fa928d4a461626e6f841b4f3/src/cache_memoize/__init__.py#L153L157

cache both the functions returns and the endpoints results: Currently only the endpoints results are cached. We use them inside the code to get quick results by taking advantage of the cache, but it's not their aim, and we have to parse / decode. It would be better to directly cache the results of the functions (memoïze). Also: we could cache the raised exceptions as here: https://github.com/peterbe/django-cache-memoize/blob/4da1ba4639774426fa928d4a461626e6f841b4f3/src/cache_memoize/__init__.py#L153L157

closed

2021-09-24T13:10:23Z

2021-10-04T16:24:45Z

2021-10-04T16:24:45Z

severo

refactor to reduce functions complexity

For example, https://github.com/huggingface/datasets-preview-backend/blob/13e533238eb6b6dfdcd8e7d3c23ed134c67b5525/src/datasets_preview_backend/queries/rows.py#L25 is rightly flagged by https://sourcery.ai/ as too convoluted. It's hard to debug and test, and there are too many special cases.

refactor to reduce functions complexity: For example, https://github.com/huggingface/datasets-preview-backend/blob/13e533238eb6b6dfdcd8e7d3c23ed134c67b5525/src/datasets_preview_backend/queries/rows.py#L25 is rightly flagged by https://sourcery.ai/ as too convoluted. It's hard to debug and test, and there are too many special cases.

closed

2021-09-24T13:06:22Z

2021-10-12T08:49:32Z

2021-10-12T08:49:31Z

severo

Add types to rows

fixes #25

Add types to rows: fixes #25

closed

2021-09-24T13:00:43Z

2021-09-24T13:06:14Z

2021-09-24T13:06:13Z

severo

"flatten" the nested values?

See https://huggingface.co/docs/datasets/process.html#flatten

"flatten" the nested values?: See https://huggingface.co/docs/datasets/process.html#flatten

closed

2021-09-24T12:58:34Z

2022-09-16T20:10:22Z

2022-09-16T20:10:22Z

severo

Info to infos

Info to infos:

closed

2021-09-24T11:26:50Z

2021-09-24T11:44:06Z

2021-09-24T11:44:05Z

severo

feat: 🎸 blocklist "allenai/c4" dataset

see https://github.com/huggingface/datasets-preview-backend/issues/17#issuecomment-918515398

feat: 🎸 blocklist "allenai/c4" dataset: see https://github.com/huggingface/datasets-preview-backend/issues/17#issuecomment-918515398

closed

2021-09-24T08:14:01Z

2021-09-24T08:18:36Z

2021-09-24T08:18:35Z

severo

use `environs` to manage the env vars?

https://pypi.org/project/environs/ instead of utils.py

use `environs` to manage the env vars?: https://pypi.org/project/environs/ instead of utils.py

closed

2021-09-24T08:05:38Z

2022-09-19T08:49:33Z

2022-09-19T08:49:33Z

severo

Grouped endpoints

Grouped endpoints:

closed

2021-09-23T18:01:49Z

2021-09-23T18:11:53Z

2021-09-23T18:11:52Z

severo

Fix serialization in benchmark

``` INFO: 127.0.0.1:38826 - "GET /info?dataset=oelkrise%2FCRT HTTP/1.1" 404 Not Found ```

Fix serialization in benchmark: ``` INFO: 127.0.0.1:38826 - "GET /info?dataset=oelkrise%2FCRT HTTP/1.1" 404 Not Found ```

closed

2021-09-23T14:57:58Z

2021-09-24T07:31:56Z

2021-09-24T07:31:56Z

severo

feat: 🎸 remove ability to chose the number of extracted rows

✅ Closes: #32

feat: 🎸 remove ability to chose the number of extracted rows: ✅ Closes: #32

closed

2021-09-23T12:10:42Z

2021-09-23T12:18:47Z

2021-09-23T12:18:46Z

severo

Move benchmark to a different repo?

It's a client of the API

Move benchmark to a different repo?: It's a client of the API

closed

2021-09-23T10:44:08Z

2021-10-12T08:49:11Z

2021-10-12T08:49:11Z

severo

Add basic cache

Fixes #3. See https://github.com/huggingface/datasets-preview-backend/milestone/1 for remaning issues.

Add basic cache: Fixes #3. See https://github.com/huggingface/datasets-preview-backend/milestone/1 for remaning issues.

closed

2021-09-23T09:54:04Z

2021-09-23T09:58:14Z

2021-09-23T09:58:14Z

severo

Enable the private datasets

The code is already present to pass the token, but it's disabled in the code (hardcoded): https://github.com/huggingface/datasets-preview-backend/blob/df04ffba9ca1a432ed65e220cf7722e518e0d4f8/src/datasets_preview_backend/cache.py#L119-L120 - [ ] enable private datasets and manage their cache adequately - [ ] separate private caches from public caches: for authenticated requests, we need to check every time, or at least use a much lower TTL, because an access can be removed. Also: since a hub dataset can be turned private, how should we manage them? - [ ] add doc. See https://github.com/huggingface/datasets-preview-backend/commit/f6576d5639bcb16fc71bf0e6796f3199124ebf49

Enable the private datasets: The code is already present to pass the token, but it's disabled in the code (hardcoded): https://github.com/huggingface/datasets-preview-backend/blob/df04ffba9ca1a432ed65e220cf7722e518e0d4f8/src/datasets_preview_backend/cache.py#L119-L120 - [ ] enable private datasets and manage their cache adequately - [ ] separate private caches from public caches: for authenticated requests, we need to check every time, or at least use a much lower TTL, because an access can be removed. Also: since a hub dataset can be turned private, how should we manage them? - [ ] add doc. See https://github.com/huggingface/datasets-preview-backend/commit/f6576d5639bcb16fc71bf0e6796f3199124ebf49

closed

2021-09-23T09:42:47Z

2024-01-31T10:14:08Z

2024-01-31T09:50:02Z

severo

Provide the ETag header

- [ ] set and manage the `ETag` header to save bandwidth when the client (browser) revalidates. See https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching and https://gist.github.com/timheap/1f4d9284e4f4d4f545439577c0ca6300 ```python # TODO: use key for ETag? It will need to be serialized # key = get_rows_json.__cache_key__( # dataset=dataset, config=config, split=split, num_rows=num_rows, token=request.user.token # ) # print(f"key={key} in cache: {cache.__contains__(key)}") ``` - [ ] ETag: add an ETag header in the response (hash of the response) - [ ] ETag: if the request contains the `If-None-Match`, parse its ETag (beware the "weak" ETags), compare to the cache, and return an empty 304 response if the cache is fresh (with or without changing the TTL), or 200 with content if it has changed

Provide the ETag header: - [ ] set and manage the `ETag` header to save bandwidth when the client (browser) revalidates. See https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching and https://gist.github.com/timheap/1f4d9284e4f4d4f545439577c0ca6300 ```python # TODO: use key for ETag? It will need to be serialized # key = get_rows_json.__cache_key__( # dataset=dataset, config=config, split=split, num_rows=num_rows, token=request.user.token # ) # print(f"key={key} in cache: {cache.__contains__(key)}") ``` - [ ] ETag: add an ETag header in the response (hash of the response) - [ ] ETag: if the request contains the `If-None-Match`, parse its ETag (beware the "weak" ETags), compare to the cache, and return an empty 304 response if the cache is fresh (with or without changing the TTL), or 200 with content if it has changed

closed

2021-09-23T09:41:40Z

2022-09-19T10:00:49Z

2022-09-19T10:00:48Z

severo

Update canonical datasets using a webhook

Webhook invalidation of canonical datasets (GitHub): - [x] setup the [`revision` argument](https://huggingface.co/docs/datasets/package_reference/loading_methods.html#datasets.load_dataset) to download datasets from the master branch - #119 - [x] set up a webhook on `datasets` library on every push to the master branch - see https://github.com/huggingface/moon-landing/issues/1345 - not needed anymore because the canonical datasets are mirrored to the hub. - [x] add an endpoint to listen to the webhook - [x] parse the webhook to find which caches should be invalidated (creation, update, deletion) - [x] refresh these caches

Update canonical datasets using a webhook: Webhook invalidation of canonical datasets (GitHub): - [x] setup the [`revision` argument](https://huggingface.co/docs/datasets/package_reference/loading_methods.html#datasets.load_dataset) to download datasets from the master branch - #119 - [x] set up a webhook on `datasets` library on every push to the master branch - see https://github.com/huggingface/moon-landing/issues/1345 - not needed anymore because the canonical datasets are mirrored to the hub. - [x] add an endpoint to listen to the webhook - [x] parse the webhook to find which caches should be invalidated (creation, update, deletion) - [x] refresh these caches

closed

2021-09-23T09:39:18Z

2022-01-26T13:44:08Z

2022-01-26T11:20:04Z

severo

Update hub datasets with webhook

Webhook invalidation of community datasets (hf.co): - [x] setup a webhook on hf.co for datasets creation, update, deletion -> waiting for https://github.com/huggingface/moon-landing/issues/1344 - [x] add an endpoint to listen to the webhook - [x] parse the webhook to find which caches should be invalidated - [x] refresh these caches - [x] document added in moonrise mongo db

Update hub datasets with webhook: Webhook invalidation of community datasets (hf.co): - [x] setup a webhook on hf.co for datasets creation, update, deletion -> waiting for https://github.com/huggingface/moon-landing/issues/1344 - [x] add an endpoint to listen to the webhook - [x] parse the webhook to find which caches should be invalidated - [x] refresh these caches - [x] document added in moonrise mongo db

closed

2021-09-23T09:38:17Z

2021-10-18T12:33:27Z

2021-10-18T12:33:27Z

severo

Refresh the cache?

Force a cache refresh on a regular basis (cron)

Refresh the cache?: Force a cache refresh on a regular basis (cron)

closed

2021-09-23T09:36:02Z

2021-10-12T08:34:41Z

2021-10-12T08:34:41Z

severo

warm the cache

Warm the cache at application startup. We want: - to avoid blocking the application, so: run asynchronously, and without hammering the server - to have a warm cache as fast as possible (persisting the previous cache, then refreshing it at startup? - related: #35 ) - [x] create a function to list all the datasets and fill the cache for all the possible requests for it. It might be `make benchmark`, or a specific function -> `make warm` - [x] persist the cache? or start with an empty cache when the application is restarted? -> yes, persisted - [x] launch it at application startup -> it's done at startup, see INSTALL.md.

warm the cache: Warm the cache at application startup. We want: - to avoid blocking the application, so: run asynchronously, and without hammering the server - to have a warm cache as fast as possible (persisting the previous cache, then refreshing it at startup? - related: #35 ) - [x] create a function to list all the datasets and fill the cache for all the possible requests for it. It might be `make benchmark`, or a specific function -> `make warm` - [x] persist the cache? or start with an empty cache when the application is restarted? -> yes, persisted - [x] launch it at application startup -> it's done at startup, see INSTALL.md.

closed

2021-09-23T09:34:35Z

2021-10-12T08:35:27Z

2021-10-12T08:35:27Z

severo

Add a parameter to specify the number of rows

It's a problem for the cache, so until we manage random access, we can: - [ ] fill the cache with a large (maximum) number of rows, ie up to 1000 - [ ] also cache the default request (N = 100) -> set to the parameter used in moon-landing - [ ] if a request comes with N = 247, for example, generate the response on the fly, from the large cache (1000), and don't cache that response

Add a parameter to specify the number of rows: It's a problem for the cache, so until we manage random access, we can: - [ ] fill the cache with a large (maximum) number of rows, ie up to 1000 - [ ] also cache the default request (N = 100) -> set to the parameter used in moon-landing - [ ] if a request comes with N = 247, for example, generate the response on the fly, from the large cache (1000), and don't cache that response

closed

2021-09-23T09:26:34Z

2022-09-16T20:13:55Z

2022-09-16T20:13:55Z

severo

Remove the `rows`/ `num_rows` argument

We will fix it to 100, in order to simplify the cache management.

Remove the `rows`/ `num_rows` argument: We will fix it to 100, in order to simplify the cache management.

closed

2021-09-23T09:24:46Z

2021-09-23T12:18:46Z

2021-09-23T12:18:46Z

severo

Manage concurrency

Currently (in the cache branch), only one worker is allowed. We want to have multiple workers, but for that we need to have a shared cache: - [ ] migrate from diskcache to redis - [ ] remove the hardcoded limit of 1 worker

Manage concurrency: Currently (in the cache branch), only one worker is allowed. We want to have multiple workers, but for that we need to have a shared cache: - [ ] migrate from diskcache to redis - [ ] remove the hardcoded limit of 1 worker

closed

2021-09-23T09:20:43Z

2021-10-06T08:08:35Z

2021-10-06T08:08:35Z

severo

Use FastAPI instead of only Starlette?

It would allow to have doc, and surely a lot of other benefits

Use FastAPI instead of only Starlette?: It would allow to have doc, and surely a lot of other benefits

closed

2021-09-17T14:45:40Z

2021-09-20T10:25:17Z

2021-09-20T07:13:00Z

severo

Ensure non-ASCII characters are handled as expected

See https://github.com/huggingface/datasets-viewer/pull/15 It should be tested

Ensure non-ASCII characters are handled as expected: See https://github.com/huggingface/datasets-viewer/pull/15 It should be tested

closed

2021-09-17T13:22:39Z

2022-09-16T20:15:19Z

2022-09-16T20:15:19Z

severo

Add endpoint to get splits + configs at once

See https://github.com/huggingface/moon-landing/pull/1040#discussion_r709494993 Also evaluate doing the same for the rows (the payload might be too heavy)

Add endpoint to get splits + configs at once: See https://github.com/huggingface/moon-landing/pull/1040#discussion_r709494993 Also evaluate doing the same for the rows (the payload might be too heavy)

closed

2021-09-16T09:14:52Z

2021-09-23T18:13:17Z

2021-09-23T18:13:17Z

severo

endpoint to generate bitmaps for mnist or cifar10 on the fly

if there are very few instances of raw image data in datasets i think it's best to generate server side vs. writing client side code no strong opinion on this though, depends on the number/variety of datasets i guess (For Audio I don't know if we have some datasets with raw tensors inside them? @lhoestq @albertvillanova )

endpoint to generate bitmaps for mnist or cifar10 on the fly: if there are very few instances of raw image data in datasets i think it's best to generate server side vs. writing client side code no strong opinion on this though, depends on the number/variety of datasets i guess (For Audio I don't know if we have some datasets with raw tensors inside them? @lhoestq @albertvillanova )

closed

2021-09-16T08:03:09Z

2021-10-18T12:42:24Z

2021-10-18T12:42:24Z

julien-c

Add endpoint to proxy local files inside datasets' data

for instance for: <img width="2012" alt="Screenshot 2021-09-16 at 09 59 38" src="https://user-images.githubusercontent.com/326577/133573786-ca9b8b60-2271-4256-b1e4-aa05302fd2f3.png">

Add endpoint to proxy local files inside datasets' data: for instance for: <img width="2012" alt="Screenshot 2021-09-16 at 09 59 38" src="https://user-images.githubusercontent.com/326577/133573786-ca9b8b60-2271-4256-b1e4-aa05302fd2f3.png">

closed

2021-09-16T08:00:11Z

2021-10-21T15:35:49Z

2021-10-21T15:35:48Z

julien-c

Add columns types to /rows response

We currently just have the keys (column identifier) and values. We might want to give each column's type: "our own serialization scheme". For ClassLabel, this means to pass the "pretty names" (the names associates to the values) along with the type See https://github.com/huggingface/moon-landing/pull/1040#discussion_r709496849

Add columns types to /rows response: We currently just have the keys (column identifier) and values. We might want to give each column's type: "our own serialization scheme". For ClassLabel, this means to pass the "pretty names" (the names associates to the values) along with the type See https://github.com/huggingface/moon-landing/pull/1040#discussion_r709496849

closed

2021-09-16T07:48:51Z

2021-09-24T13:06:13Z

2021-09-24T13:06:13Z

severo

Expose a list of "valid" i.e. previewable datasets for moon-landing to be able to tag/showcase them

(linked to caching and pre-warming, obviously)

Expose a list of "valid" i.e. previewable datasets for moon-landing to be able to tag/showcase them: (linked to caching and pre-warming, obviously)

closed

2021-09-15T19:28:47Z

2021-10-04T19:51:55Z

2021-10-04T19:51:55Z

julien-c

run benchmark automatically every week, and store the results

- [ ] create a github action? - [ ] store the report in... an HF dataset? or in a github repo? should it be private? - [ ] get these reports from https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading

run benchmark automatically every week, and store the results: - [ ] create a github action? - [ ] store the report in... an HF dataset? or in a github repo? should it be private? - [ ] get these reports from https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading

closed

2021-09-15T12:15:00Z

2021-10-12T08:48:02Z

2021-10-12T08:48:01Z

severo

`make benchmark` is very long and blocks

Sometimes `make benchmark` blocks (nothing happens, and only one process is running, while the load is low). Ideally, it would not block, and other processes would be launched anyway so that the full capacity of the CPUs would be used (`-j -l 7` parameters of `make`) To unblock, I have to kill and relaunch `make benchmark` manually.

`make benchmark` is very long and blocks: Sometimes `make benchmark` blocks (nothing happens, and only one process is running, while the load is low). Ideally, it would not block, and other processes would be launched anyway so that the full capacity of the CPUs would be used (`-j -l 7` parameters of `make`) To unblock, I have to kill and relaunch `make benchmark` manually.

closed

2021-09-14T07:42:44Z

2021-10-12T08:34:17Z

2021-10-12T08:34:17Z

severo

exception seen during `make benchmark`

Not sure which dataset threw this exception though, that's why I put the previous rows for further investigation. ``` poetry run python ../scripts/get_rows_report.py wikiann___CONFIG___or___SPLIT___test ../tmp/get_rows_reports/wikiann___CONFIG___or___SPLIT___test.json poetry run python ../scripts/get_rows_report.py csebuetnlp___SLASH___xlsum___CONFIG___uzbek___SPLIT___test ../tmp/get_rows_reports/csebuetnlp___SLASH___xlsum___CONFIG___uzbek___SPLIT___test.json poetry run python ../scripts/get_rows_report.py clips___SLASH___mfaq___CONFIG___no___SPLIT___train ../tmp/get_rows_reports/clips___SLASH___mfaq___CONFIG___no___SPLIT___train.json poetry run python ../scripts/get_rows_report.py common_voice___CONFIG___rm-vallader___SPLIT___train ../tmp/get_rows_reports/common_voice___CONFIG___rm-vallader___SPLIT___train.json https://media.githubusercontent.com/media/persiannlp/parsinlu/master/data/translation/translation_combined_fa_en/test.tsv poetry run python ../scripts/get_rows_report.py pasinit___SLASH___xlwic___CONFIG___xlwic_en_da___SPLIT___train ../tmp/get_rows_reports/pasinit___SLASH___xlwic___CONFIG___xlwic_en_da___SPLIT___train.json poetry run python ../scripts/get_rows_report.py indic_glue___CONFIG___wstp.mr___SPLIT___validation ../tmp/get_rows_reports/indic_glue___CONFIG___wstp.mr___SPLIT___validation.json poetry run python ../scripts/get_rows_report.py banking77___CONFIG___default___SPLIT___test ../tmp/get_rows_reports/banking77___CONFIG___default___SPLIT___test.json poetry run python ../scripts/get_rows_report.py gem___CONFIG___xsum___SPLIT___challenge_test_bfp_05 ../tmp/get_rows_reports/gem___CONFIG___xsum___SPLIT___challenge_test_bfp_05.json poetry run python ../scripts/get_rows_report.py turingbench___SLASH___TuringBench___CONFIG___TT_fair_wmt19___SPLIT___validation ../tmp/get_rows_reports/turingbench___SLASH___TuringBench___CONFIG___TT_fair_wmt19___SPLIT___validation.json poetry run python ../scripts/get_rows_report.py igbo_monolingual___CONFIG___eze_goes_to_school___SPLIT___train ../tmp/get_rows_reports/igbo_monolingual___CONFIG___eze_goes_to_school___SPLIT___train.json poetry run python ../scripts/get_rows_report.py flax-sentence-embeddings___SLASH___stackexchange_titlebody_best_voted_answer_jsonl___CONFIG___gamedev___SPLIT___train ../tmp/get_rows_reports/flax-sentence-embeddings___SLASH___stackexchange_titlebody_best_voted_answer_jsonl___CONFIG___gamedev___SPLIT___train.json * skipping . . . Exception ignored in: <generator object ParsinluReadingComprehension._generate_examples at 0x7f094caa6dd0> Traceback (most recent call last): File "/home/slesage/hf/datasets-preview-backend/.venv/lib/python3.8/site-packages/datasets/iterable_dataset.py", line 79, in __iter__ yield key, example RuntimeError: generator ignored GeneratorExit ```

exception seen during `make benchmark`: Not sure which dataset threw this exception though, that's why I put the previous rows for further investigation. ``` poetry run python ../scripts/get_rows_report.py wikiann___CONFIG___or___SPLIT___test ../tmp/get_rows_reports/wikiann___CONFIG___or___SPLIT___test.json poetry run python ../scripts/get_rows_report.py csebuetnlp___SLASH___xlsum___CONFIG___uzbek___SPLIT___test ../tmp/get_rows_reports/csebuetnlp___SLASH___xlsum___CONFIG___uzbek___SPLIT___test.json poetry run python ../scripts/get_rows_report.py clips___SLASH___mfaq___CONFIG___no___SPLIT___train ../tmp/get_rows_reports/clips___SLASH___mfaq___CONFIG___no___SPLIT___train.json poetry run python ../scripts/get_rows_report.py common_voice___CONFIG___rm-vallader___SPLIT___train ../tmp/get_rows_reports/common_voice___CONFIG___rm-vallader___SPLIT___train.json https://media.githubusercontent.com/media/persiannlp/parsinlu/master/data/translation/translation_combined_fa_en/test.tsv poetry run python ../scripts/get_rows_report.py pasinit___SLASH___xlwic___CONFIG___xlwic_en_da___SPLIT___train ../tmp/get_rows_reports/pasinit___SLASH___xlwic___CONFIG___xlwic_en_da___SPLIT___train.json poetry run python ../scripts/get_rows_report.py indic_glue___CONFIG___wstp.mr___SPLIT___validation ../tmp/get_rows_reports/indic_glue___CONFIG___wstp.mr___SPLIT___validation.json poetry run python ../scripts/get_rows_report.py banking77___CONFIG___default___SPLIT___test ../tmp/get_rows_reports/banking77___CONFIG___default___SPLIT___test.json poetry run python ../scripts/get_rows_report.py gem___CONFIG___xsum___SPLIT___challenge_test_bfp_05 ../tmp/get_rows_reports/gem___CONFIG___xsum___SPLIT___challenge_test_bfp_05.json poetry run python ../scripts/get_rows_report.py turingbench___SLASH___TuringBench___CONFIG___TT_fair_wmt19___SPLIT___validation ../tmp/get_rows_reports/turingbench___SLASH___TuringBench___CONFIG___TT_fair_wmt19___SPLIT___validation.json poetry run python ../scripts/get_rows_report.py igbo_monolingual___CONFIG___eze_goes_to_school___SPLIT___train ../tmp/get_rows_reports/igbo_monolingual___CONFIG___eze_goes_to_school___SPLIT___train.json poetry run python ../scripts/get_rows_report.py flax-sentence-embeddings___SLASH___stackexchange_titlebody_best_voted_answer_jsonl___CONFIG___gamedev___SPLIT___train ../tmp/get_rows_reports/flax-sentence-embeddings___SLASH___stackexchange_titlebody_best_voted_answer_jsonl___CONFIG___gamedev___SPLIT___train.json * skipping . . . Exception ignored in: <generator object ParsinluReadingComprehension._generate_examples at 0x7f094caa6dd0> Traceback (most recent call last): File "/home/slesage/hf/datasets-preview-backend/.venv/lib/python3.8/site-packages/datasets/iterable_dataset.py", line 79, in __iter__ yield key, example RuntimeError: generator ignored GeneratorExit ```

closed

2021-09-14T07:33:59Z

2021-09-23T12:39:13Z

2021-09-23T12:39:13Z

severo

Upgrade datasets to 1.12.0

- [x] See https://github.com/huggingface/datasets/releases/tag/1.12.0 - [x] launch benchmark and report to https://github.com/huggingface/datasets-preview-backend/issues/9

Upgrade datasets to 1.12.0: - [x] See https://github.com/huggingface/datasets/releases/tag/1.12.0 - [x] launch benchmark and report to https://github.com/huggingface/datasets-preview-backend/issues/9

closed

2021-09-13T18:49:24Z

2021-09-14T08:31:24Z

2021-09-14T08:31:24Z

severo

Add unit tests to CI

Add unit tests to CI:

closed

2021-09-01T12:30:48Z

2021-09-02T12:56:45Z

2021-09-02T12:56:45Z

severo

CI: how to acknowledge a "safety" warning?

We use `safety` to check vulnerabilities in the dependencies. But in the case below, `tensorflow` is marked as insecure while the last published version on pipy is still 2.6.0. What to do in this case? ``` +==============================================================================+ | | | /$$$$$$ /$$ | | /$$__ $$ | $$ | | /$$$$$$$ /$$$$$$ | $$ \__//$$$$$$ /$$$$$$ /$$ /$$ | | /$$_____/ |____ $$| $$$$ /$$__ $$|_ $$_/ | $$ | $$ | | | $$$$$$ /$$$$$$$| $$_/ | $$$$$$$$ | $$ | $$ | $$ | | \____ $$ /$$__ $$| $$ | $$_____/ | $$ /$$| $$ | $$ | | /$$$$$$$/| $$$$$$$| $$ | $$$$$$$ | $$$$/| $$$$$$$ | | |_______/ \_______/|__/ \_______/ \___/ \____ $$ | | /$$ | $$ | | | $$$$$$/ | | by pyup.io \______/ | | | +==============================================================================+ | REPORT | | checked 137 packages, using free DB (updated once a month) | +============================+===========+==========================+==========+ | package | installed | affected | ID | +============================+===========+==========================+==========+ | tensorflow | 2.6.0 | ==2.6.0 | 41161 | +==============================================================================+ ```

CI: how to acknowledge a "safety" warning?: We use `safety` to check vulnerabilities in the dependencies. But in the case below, `tensorflow` is marked as insecure while the last published version on pipy is still 2.6.0. What to do in this case? ``` +==============================================================================+ | | | /$$$$$$ /$$ | | /$$__ $$ | $$ | | /$$$$$$$ /$$$$$$ | $$ \__//$$$$$$ /$$$$$$ /$$ /$$ | | /$$_____/ |____ $$| $$$$ /$$__ $$|_ $$_/ | $$ | $$ | | | $$$$$$ /$$$$$$$| $$_/ | $$$$$$$$ | $$ | $$ | $$ | | \____ $$ /$$__ $$| $$ | $$_____/ | $$ /$$| $$ | $$ | | /$$$$$$$/| $$$$$$$| $$ | $$$$$$$ | $$$$/| $$$$$$$ | | |_______/ \_______/|__/ \_______/ \___/ \____ $$ | | /$$ | $$ | | | $$$$$$/ | | by pyup.io \______/ | | | +==============================================================================+ | REPORT | | checked 137 packages, using free DB (updated once a month) | +============================+===========+==========================+==========+ | package | installed | affected | ID | +============================+===========+==========================+==========+ | tensorflow | 2.6.0 | ==2.6.0 | 41161 | +==============================================================================+ ```

closed

2021-09-01T07:20:45Z

2021-09-15T11:58:56Z

2021-09-15T11:58:48Z

severo

Prevent DoS when accessing some datasets

For example: https://huggingface.co/datasets/allenai/c4 script is doing 69219 output requests **on every received request**, which occupies all the CPUs. ``` pm2 logs ``` ``` 0|datasets | INFO: 3.238.194.17:0 - "GET /configs?dataset=allenai/c4 HTTP/1.1" 200 OK ``` ``` Check remote data files: 78%|███████▊ | 54330/69219 [14:13<03:05, 80.10it/s] Check remote data files: 79%|███████▊ | 54349/69219 [14:13<03:51, 64.14it/s] Check remote data files: 79%|███████▊ | 54364/69219 [14:14<04:44, 52.14it/s] Check remote data files: 79%|███████▊ | 54375/69219 [14:14<04:48, 51.38it/s] Check remote data files: 79%|███████▊ | 54448/69219 [14:15<02:37, 93.81it/s] Check remote data files: 79%|███████▉ | 54543/69219 [14:15<01:56, 125.60it/s] Check remote data files: 79%|███████▉ | 54564/69219 [14:16<03:22, 72.33it/s] ```

Prevent DoS when accessing some datasets: For example: https://huggingface.co/datasets/allenai/c4 script is doing 69219 output requests **on every received request**, which occupies all the CPUs. ``` pm2 logs ``` ``` 0|datasets | INFO: 3.238.194.17:0 - "GET /configs?dataset=allenai/c4 HTTP/1.1" 200 OK ``` ``` Check remote data files: 78%|███████▊ | 54330/69219 [14:13<03:05, 80.10it/s] Check remote data files: 79%|███████▊ | 54349/69219 [14:13<03:51, 64.14it/s] Check remote data files: 79%|███████▊ | 54364/69219 [14:14<04:44, 52.14it/s] Check remote data files: 79%|███████▊ | 54375/69219 [14:14<04:48, 51.38it/s] Check remote data files: 79%|███████▊ | 54448/69219 [14:15<02:37, 93.81it/s] Check remote data files: 79%|███████▉ | 54543/69219 [14:15<01:56, 125.60it/s] Check remote data files: 79%|███████▉ | 54564/69219 [14:16<03:22, 72.33it/s] ```

closed

2021-08-31T15:56:11Z

2021-10-15T15:57:49Z

2021-10-15T15:57:49Z

severo

Raise an issue when no row can be fetched

Currently, https://datasets-preview.huggingface.tech/rows?dataset=superb&config=asr&split=train&rows=5 returns ``` { "dataset": "superb", "config": "asr", "split": "train", "rows": [ ] } ``` while it should return 5 rows. An error should be raised in that case. Beware: manage the special case when the query parameter `rows` is greater than the number of rows in the split. In that case, it's normal that the number of returned rows is lower than requested.

Raise an issue when no row can be fetched: Currently, https://datasets-preview.huggingface.tech/rows?dataset=superb&config=asr&split=train&rows=5 returns ``` { "dataset": "superb", "config": "asr", "split": "train", "rows": [ ] } ``` while it should return 5 rows. An error should be raised in that case. Beware: manage the special case when the query parameter `rows` is greater than the number of rows in the split. In that case, it's normal that the number of returned rows is lower than requested.

closed

2021-08-27T12:36:50Z

2021-09-14T08:50:53Z

2021-09-14T08:50:53Z

severo

Add an endpoint to get the dataset card?

See https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/hf_api.py#L427, `full` argument The dataset card is the README.md.

Add an endpoint to get the dataset card?: See https://github.com/huggingface/huggingface_hub/blob/main/src/huggingface_hub/hf_api.py#L427, `full` argument The dataset card is the README.md.

closed

2021-08-26T13:43:29Z

2022-09-16T20:15:52Z

2022-09-16T20:15:52Z

severo

Properly manage the case config is None

For example: https://datasets-preview.huggingface.tech/splits?dataset=sent_comp&config=null returns ~~~json { "dataset": "sent_comp", "config": "null", "splits": [ "validation", "train" ] } ~~~ this should have errored since there is no `"null"` config (it's `null`). https://datasets-preview.huggingface.tech/splits?dataset=sent_comp&config= returns ~~~json The split names could not be parsed from the dataset config. ~~~ https://datasets-preview.huggingface.tech/splits?dataset=sent_comp returns ~~~json { "dataset": "sent_comp", "config": null, "splits": [ "validation", "train" ] } ~~~ --- As a reference for the same dataset https://datasets-preview.huggingface.tech/configs?dataset=sent_comp returns ~~~json { "dataset": "sent_comp", "configs": [ null ] } ~~~ and https://datasets-preview.huggingface.tech/info?dataset=sent_comp returns ~~~json { "dataset": "sent_comp", "info": { "default": { ... "builder_name": "sent_comp", "config_name": "default", ... "splits": { "validation": { "name": "validation", "num_bytes": 55823979, "num_examples": 10000, "dataset_name": "sent_comp" }, "train": { "name": "train", "num_bytes": 1135684803, "num_examples": 200000, "dataset_name": "sent_comp" } }, ... } } } ~~~

Properly manage the case config is None: For example: https://datasets-preview.huggingface.tech/splits?dataset=sent_comp&config=null returns ~~~json { "dataset": "sent_comp", "config": "null", "splits": [ "validation", "train" ] } ~~~ this should have errored since there is no `"null"` config (it's `null`). https://datasets-preview.huggingface.tech/splits?dataset=sent_comp&config= returns ~~~json The split names could not be parsed from the dataset config. ~~~ https://datasets-preview.huggingface.tech/splits?dataset=sent_comp returns ~~~json { "dataset": "sent_comp", "config": null, "splits": [ "validation", "train" ] } ~~~ --- As a reference for the same dataset https://datasets-preview.huggingface.tech/configs?dataset=sent_comp returns ~~~json { "dataset": "sent_comp", "configs": [ null ] } ~~~ and https://datasets-preview.huggingface.tech/info?dataset=sent_comp returns ~~~json { "dataset": "sent_comp", "info": { "default": { ... "builder_name": "sent_comp", "config_name": "default", ... "splits": { "validation": { "name": "validation", "num_bytes": 55823979, "num_examples": 10000, "dataset_name": "sent_comp" }, "train": { "name": "train", "num_bytes": 1135684803, "num_examples": 200000, "dataset_name": "sent_comp" } }, ... } } } ~~~

closed

2021-08-26T12:16:27Z

2021-08-26T13:26:00Z

2021-08-26T13:26:00Z

severo

Get random access to the rows

Currently, only the first rows can be obtained with /rows. We want to get access to slices of the rows through pagination, eg /rows?from=40000&rows=10

Get random access to the rows: Currently, only the first rows can be obtained with /rows. We want to get access to slices of the rows through pagination, eg /rows?from=40000&rows=10

closed

2021-08-26T08:21:34Z

2023-06-14T12:16:22Z

2023-06-14T12:16:22Z

severo

Install the datasets that require manual download

Some datasets require a manual download (https://huggingface.co/datasets/arxiv_dataset, for example). We might manually download them on the server, so that the backend returns the rows, instead of an error.

Install the datasets that require manual download: Some datasets require a manual download (https://huggingface.co/datasets/arxiv_dataset, for example). We might manually download them on the server, so that the backend returns the rows, instead of an error.

closed

2021-08-25T16:30:11Z

2022-06-17T11:47:18Z

2022-06-17T11:47:18Z

severo

Give the cause of the error in the endpoints

It can thus be used in the hub to show hints to the dataset owner (or user?) to improve the script and fix the bug.

Give the cause of the error in the endpoints: It can thus be used in the hub to show hints to the dataset owner (or user?) to improve the script and fix the bug.

closed

2021-08-25T09:45:58Z

2021-08-26T14:39:22Z

2021-08-26T14:39:21Z

severo

Use /info as the source for configs and splits?

It's a refactor. As the dataset info contains the configs and splits, maybe the code can be factorized. Before doing it: review the errors for /info, /configs, and /splits (https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading) and ensure we will not increase the number of erroneous datasets.

Use /info as the source for configs and splits?: It's a refactor. As the dataset info contains the configs and splits, maybe the code can be factorized. Before doing it: review the errors for /info, /configs, and /splits (https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading) and ensure we will not increase the number of erroneous datasets.

closed

2021-08-25T09:43:51Z

2021-09-01T07:08:25Z

2021-09-01T07:08:25Z

severo

Increase the proportion of hf.co datasets that can be previewed

For different reasons, some datasets cannot be previewed. It might be because the loading script is buggy, because the data is in a format that cannot be streamed, etc. The script https://github.com/huggingface/datasets-preview-backend/blob/master/quality/test_datasets.py tests the three endpoints on all the datasets in hf.co and outputs a data file that can be analysed in https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading. The goal is to understand which problems arise most and try to fix the ones that can be addressed (here or in [datasets](https://github.com/huggingface/datasets/)) so that the largest part of the hf.co datasets can be previewed.

Increase the proportion of hf.co datasets that can be previewed: For different reasons, some datasets cannot be previewed. It might be because the loading script is buggy, because the data is in a format that cannot be streamed, etc. The script https://github.com/huggingface/datasets-preview-backend/blob/master/quality/test_datasets.py tests the three endpoints on all the datasets in hf.co and outputs a data file that can be analysed in https://observablehq.com/@huggingface/quality-assessment-of-datasets-loading. The goal is to understand which problems arise most and try to fix the ones that can be addressed (here or in [datasets](https://github.com/huggingface/datasets/)) so that the largest part of the hf.co datasets can be previewed.

closed

2021-08-17T10:17:23Z

2022-01-31T21:34:27Z

2022-01-31T21:34:07Z

severo

Add CI

Check types and code quality

Add CI: Check types and code quality

closed

2021-08-17T10:09:35Z

2021-09-01T07:09:56Z

2021-09-01T07:09:56Z

severo

Support private datasets

For now, only public datasets can be queried. To support private datasets : - [x] add `use_auth_token` argument to all the queries functions (and upstream too in https://github.com/huggingface/datasets/blob/master/src/datasets/inspect.py) - [x] obtain the authentication header <strike>or cookie</strike> from the request

Support private datasets: For now, only public datasets can be queried. To support private datasets : - [x] add `use_auth_token` argument to all the queries functions (and upstream too in https://github.com/huggingface/datasets/blob/master/src/datasets/inspect.py) - [x] obtain the authentication header <strike>or cookie</strike> from the request

closed

2021-08-09T14:19:51Z

2021-09-16T08:27:33Z

2021-09-16T08:27:32Z

severo

Expand the purpose of this backend?

Depending on the evolution of https://github.com/huggingface/datasets, this project might disappear, or its features might be reduced, in particular, if one day it allows caching the data by self-generating: - an arrow or a parquet data file (maybe with sharding and compression for the largest datasets) - or a SQL database - or precompute and store a partial list of known offsets (every 10MB for example) It would allow getting random access to the data.

Expand the purpose of this backend?: Depending on the evolution of https://github.com/huggingface/datasets, this project might disappear, or its features might be reduced, in particular, if one day it allows caching the data by self-generating: - an arrow or a parquet data file (maybe with sharding and compression for the largest datasets) - or a SQL database - or precompute and store a partial list of known offsets (every 10MB for example) It would allow getting random access to the data.

closed

2021-08-09T14:03:41Z

2022-02-04T11:24:32Z

2022-02-04T11:24:32Z

severo

Upgrade `datasets` and adapt the tests

Two issues have been fixed in `datasets`: - https://github.com/huggingface/datasets/issues/2743 - https://github.com/huggingface/datasets/issues/2749 Also, support for streaming compressed files is improving: - https://github.com/huggingface/datasets/pull/2786 - https://github.com/huggingface/datasets/pull/2800 On the next `datasets` release (see https://github.com/huggingface/datasets/releases), upgrade the dependency here and change the integration tests.

Upgrade `datasets` and adapt the tests: Two issues have been fixed in `datasets`: - https://github.com/huggingface/datasets/issues/2743 - https://github.com/huggingface/datasets/issues/2749 Also, support for streaming compressed files is improving: - https://github.com/huggingface/datasets/pull/2786 - https://github.com/huggingface/datasets/pull/2800 On the next `datasets` release (see https://github.com/huggingface/datasets/releases), upgrade the dependency here and change the integration tests.

closed

2021-08-09T09:08:33Z

2021-09-01T07:09:31Z

2021-09-01T07:09:31Z

severo

Establish and meet SLO

https://en.wikipedia.org/wiki/Service-level_objective as stated in https://github.com/huggingface/datasets-preview-backend/issues/1#issuecomment-894430211: > we need to "guarantee" that row fetches from moon-landing will be under a specified latency (to be discussed), even in the case of cache misses in `datasets-preview-backend` > > because the data will be needed at server-rendering time, for content to be parsed by Google > > What's a reasonable latency you think you can achieve? > > If it's too long we might want to pre-warm the cache for all (streamable) dataset, using a system based on webhooks from moon-landing for instance See also https://github.com/huggingface/datasets-preview-backend/issues/3 for the cache.

Establish and meet SLO: https://en.wikipedia.org/wiki/Service-level_objective as stated in https://github.com/huggingface/datasets-preview-backend/issues/1#issuecomment-894430211: > we need to "guarantee" that row fetches from moon-landing will be under a specified latency (to be discussed), even in the case of cache misses in `datasets-preview-backend` > > because the data will be needed at server-rendering time, for content to be parsed by Google > > What's a reasonable latency you think you can achieve? > > If it's too long we might want to pre-warm the cache for all (streamable) dataset, using a system based on webhooks from moon-landing for instance See also https://github.com/huggingface/datasets-preview-backend/issues/3 for the cache.

closed

2021-08-09T08:47:22Z

2022-09-07T15:21:13Z

2022-09-07T15:21:12Z

severo

Cache the responses

The datasets generally don't change often, so it's surely worth caching the responses. Three levels of cache are involved: - client (browser, moon-landing): use Response headers (cache-control, ETag, see https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching) - application: serve the cached responses. Invalidation of the cache for a given request: - when the request arrives, if the TTL has finished - when a webhook has been received for the dataset (see below) - (needed?) when a dedicated background process is launched to refresh the cache (cron) - `datasets`: the library manages [its own cache](https://huggingface.co/docs/datasets/processing.html#controlling-the-cache-behavior) to avoid unneeded downloads Here we will implement the application cache, and provide the headers for the client cache. - [x] cache the responses (content and status code) during a TTL - [x] select a cache library: - -> http://www.grantjenks.com/docs/diskcache/tutorial.html - redis (https://redis.io/topics/persistence) - https://github.com/florimondmanca/asgi-caches - [x] check the size of the cache and allocate sufficient resources. Note that every request generates a very small JSON (in the worst case, it's the dataset-info.json file, for ~3,000 datasets, else it's a JSON with at most some strings). The only problem would be if we're flooded by random requests (which generate 404 errors and are cached). Anyway, there is a limit to 1GB (the default in diskcache) - [x] generate a key from the request - [x] store and retrieve the response - [x] specify the TTL - [x] configure the TTL as an option - [x] set the `cache-control` header so that the client (browser) doesn't retry during some time. See https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching - [x] TTL: return "max-age" in the `Cache-Control` header (computed based on the server TTL to match the same date? or set `Expires` instead?) - [x] <strike>Manage `If-Modified-Since` header in request?</strike>: no, this header works with `Last-Modified`, not `Cache-Control`/ `max-age` - [x] manage concurrency - [x] allow launching various workers. Done with `WEB_CONCURRENCY` - currently hardcoded to 1 - [x] <strike>migrate to Redis - the cache service will be separated from the application</strike> -> moved to a dedicated issue: #31 The scope of this issue has been reduced. See the next issues: - #32 - #31 - #34 - #36 - #37 And maybe: - #35 - #38 - #39

Cache the responses: The datasets generally don't change often, so it's surely worth caching the responses. Three levels of cache are involved: - client (browser, moon-landing): use Response headers (cache-control, ETag, see https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching) - application: serve the cached responses. Invalidation of the cache for a given request: - when the request arrives, if the TTL has finished - when a webhook has been received for the dataset (see below) - (needed?) when a dedicated background process is launched to refresh the cache (cron) - `datasets`: the library manages [its own cache](https://huggingface.co/docs/datasets/processing.html#controlling-the-cache-behavior) to avoid unneeded downloads Here we will implement the application cache, and provide the headers for the client cache. - [x] cache the responses (content and status code) during a TTL - [x] select a cache library: - -> http://www.grantjenks.com/docs/diskcache/tutorial.html - redis (https://redis.io/topics/persistence) - https://github.com/florimondmanca/asgi-caches - [x] check the size of the cache and allocate sufficient resources. Note that every request generates a very small JSON (in the worst case, it's the dataset-info.json file, for ~3,000 datasets, else it's a JSON with at most some strings). The only problem would be if we're flooded by random requests (which generate 404 errors and are cached). Anyway, there is a limit to 1GB (the default in diskcache) - [x] generate a key from the request - [x] store and retrieve the response - [x] specify the TTL - [x] configure the TTL as an option - [x] set the `cache-control` header so that the client (browser) doesn't retry during some time. See https://developer.mozilla.org/en-US/docs/Web/HTTP/Caching - [x] TTL: return "max-age" in the `Cache-Control` header (computed based on the server TTL to match the same date? or set `Expires` instead?) - [x] <strike>Manage `If-Modified-Since` header in request?</strike>: no, this header works with `Last-Modified`, not `Cache-Control`/ `max-age` - [x] manage concurrency - [x] allow launching various workers. Done with `WEB_CONCURRENCY` - currently hardcoded to 1 - [x] <strike>migrate to Redis - the cache service will be separated from the application</strike> -> moved to a dedicated issue: #31 The scope of this issue has been reduced. See the next issues: - #32 - #31 - #34 - #36 - #37 And maybe: - #35 - #38 - #39

closed

2021-08-03T14:38:12Z

2021-09-23T09:58:13Z

2021-09-23T09:58:13Z

severo

Instrument the application

Measure the response time, status code, RAM usage, etc. to be able to take decision (see #1). Also statistics about the most common requests (endpoint, dataset, parameters)

Instrument the application: Measure the response time, status code, RAM usage, etc. to be able to take decision (see #1). Also statistics about the most common requests (endpoint, dataset, parameters)

closed

2021-08-03T14:32:10Z

2022-09-16T20:16:46Z

2022-09-16T20:16:46Z

severo

Scale the application

Both `uvicorn` and `pm2` allow specifying the number of workers. `pm2` seems interesting since it provides a way to increase or decrease the number of workers without restart. But before using multiple workers, it's important to instrument the app in order to detect if we need it (eg: monitor the response time).

Scale the application: Both `uvicorn` and `pm2` allow specifying the number of workers. `pm2` seems interesting since it provides a way to increase or decrease the number of workers without restart. But before using multiple workers, it's important to instrument the app in order to detect if we need it (eg: monitor the response time).

closed

2021-08-03T14:31:31Z

2022-05-11T15:09:59Z

2022-05-11T15:09:59Z

severo