The Hugging Face component is an AI component that allows users to connect the AI models served on the Hugging Face Platform.
It can carry out the following tasks:
The component definition and tasks are defined in the definition.yaml and tasks.yaml files respectively.
Setup
In order to communicate with Hugging Face, the following connection details need to be
provided. You may specify them directly in a pipeline recipe as key-value pairs
within the component's setup block, or you can create a Connection from
the Integration Settings
page and reference the whole setup as setup: ${connection.<my-connection-id>}.
Field
Field ID
Type
Note
API Key (required)
api-key
string
Fill in your Hugging face API token. To find your token, visit here.
Base URL (required)
base-url
string
Hostname for the endpoint. To use Inference API set to here, for Inference Endpoint set to your custom endpoint.
Is Custom Endpoint (required)
is-custom-endpoint
boolean
Fill true if you are using a custom Inference Endpoint and not the Inference API.
Supported Tasks
Text Generation
Generating text is the task of producing new text. These models can, for example, fill in incomplete text or paraphrase.
Whether or not to use sampling, use greedy decoding otherwise.
Max New Tokens
max-new-tokens
integer
The amount of new tokens to be generated, this does not include the input length it is a estimate of the size of generated text you want. Each new tokens slows down the request, so look for balance between response times and length of text generated.
Max Time
max-time
number
The amount of time in seconds that the query should take maximum. Network can cause some overhead so it will be a soft limit. Use that in combination with max-new-tokens for best results.
Num Return Sequences
num-return-sequences
integer
The number of proposition you want to be returned.
Repetition Penalty
repetition-penalty
number
The more a token is used within generation the more it is penalized to not be picked in successive generation passes.
Return Full Text
return-full-text
boolean
If set to False, the return results will not contain the original query making it easier for prompting.
Temperature
temperature
number
The temperature of the sampling operation. 1 means regular sampling, 0 means always take the highest score, 100.0 is getting closer to uniform probability.
Top K
top-k
integer
Integer to define the top tokens considered within the sample operation to create new text.
Top P
top-p
number
Float to define the tokens that are within the sample operation of text generation. Add tokens in the sample for more probable to least probable until the sum of the probabilities is greater than top-p.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Generated Text
generated-text
string
The continuated string.
Fill Mask
Masked language modeling is the task of masking some of the words in a sentence and predicting which words should replace those masks. These models are useful when we want to get a statistical understanding of the language in which the model is trained in.
Input
Field ID
Type
Description
Task ID (required)
task
string
TASK_FILL_MASK
Model (required)
model
string
The Hugging Face model to be used.
String Input (required)
inputs
string
a string to be filled from, must contain the [MASK] token (check model card for exact name of the mask).
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
The actual sequence of tokens that ran against the model (may contain special tokens).
Token
token
integer
The id of the token.
Token Str
token-str
string
The string representation of the token.
Summarization
Summarization is the task of producing a shorter version of a document while preserving its important information. Some models can extract text from the original input, while other models can generate entirely new text.
Integer to define the maximum length in tokens of the output summary.
Max Time
max-time
number
The amount of time in seconds that the query should take maximum. Network can cause some overhead so it will be a soft limit.
Min Length
min-length
integer
Integer to define the minimum length in tokens of the output summary.
Repetition Penalty
repetition-penalty
number
The more a token is used within generation the more it is penalized to not be picked in successive generation passes.
Temperature
temperature
number
The temperature of the sampling operation. 1 means regular sampling, 0 means always take the highest score, 100.0 is getting closer to uniform probability.
Top K
top-k
integer
Integer to define the top tokens considered within the sample operation to create new text.
Top P
top-p
number
Float to define the tokens that are within the sample operation of text generation. Add tokens in the sample for more probable to least probable until the sum of the probabilities is greater than top-p.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Summary Text
summary-text
string
The string after summarization.
Text Classification
Text Classification is the task of assigning a label or class to a given text. Some use cases are sentiment analysis, natural language inference, and assessing grammatical correctness.
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
A floats that represents how likely is that the text belongs the this class.
Token Classification
Token classification is a natural language understanding task in which a label is assigned to some tokens in a text. Some popular token classification subtasks are Named Entity Recognition (NER) and Part-of-Speech (PoS) tagging. NER models could be trained to identify specific entities in a text, such as dates, individuals and places; and PoS tagging would identify, for example, which words in a text are verbs, nouns, and punctuation marks.
none: Every token gets classified without further aggregation.
simple: Entities are grouped according to the default schema (B-, I- tags get merged when the tag is similar).
first: Same as the simple strategy except words cannot end up with different tags. Words will use the tag of the first token when there is ambiguity.
average: Same as the simple strategy except words cannot end up with different tags. Scores are averaged across tokens and then the maximum label is applied.
max: Same as the simple strategy except words cannot end up with different tags. Word entity will be the token with the maximum score.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Translation Text
translation-text
string
The string after translation.
Zero Shot Classification
Zero-shot text classification is a task in natural language processing where a model is trained on a set of labeled examples but is then able to classify new examples from previously unseen classes.
a list of strings that are potential classes for inputs. (max 10 candidate-labels, for more, simply run multiple requests, results are going to be misleading if using too many candidate-labels anyway. If you want to keep the exact same, you can simply run multi-label=True and do the scaling on your end. ).
Multi Label
multi-label
boolean
Boolean that is set to True if classes can overlap.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Scores
scores
array[number]
a list of floats that correspond the the probability of label, in the same order as labels.
Labels
labels
array[string]
The list of strings for labels that you sent (in order).
Sequence (optional)
sequence
string
The string sent as an input.
Question Answering
Question Answering models can retrieve the answer to a question from a given text, which is useful for searching for an answer in a document. Some question answering models can generate answers without context!.
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Answer
answer
string
A string that’s the answer within the text.
Stop (optional)
stop
integer
The index (string wise) of the stop of the answer within context.
Score (optional)
score
number
A float that represents how likely that the answer is correct.
Start (optional)
start
integer
The index (string wise) of the start of the answer within context.
Table Question Answering
Table Question Answering (Table QA) is the answering a question about an information on a given table.
The query in plain text that you want to ask the table.
Table
table
object
A table of data represented as a dict of list where entries are headers and the lists are all the values, all lists must have the same size.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Aggregator (optional)
aggregator
string
The aggregator used to get the answer.
Answer
answer
string
The plaintext answer.
Cells (optional)
cells
array[string]
a list of coordinates of the cells contents.
Coordinates (optional)
coordinates
array[array]
a list of coordinates of the cells referenced in the answer.
Sentence Similarity
Sentence Similarity is the task of determining how similar two texts are. Sentence similarity models convert input texts into vectors (embeddings) that capture semantic information and calculate how close (similar) they are between them. This task is particularly useful for information retrieval and clustering/grouping.
A list of strings which will be compared against the source-sentence.
Source Sentence
source-sentence
string
The string that you wish to compare the other strings with. This can be a phrase, sentence, or longer passage, depending on the model being used.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
Output
Field ID
Type
Description
Scores
scores
array[number]
The associated similarity score for each of the given strings.
Conversational
Conversational response modelling is the task of generating conversational text that is relevant, coherent and knowledgable given a prompt. These models have applications in chatbots, and as a part of voice assistants.
A list of strings corresponding to the earlier replies from the model.
Past User Inputs
past-user-inputs
array
A list of strings corresponding to the earlier replies from the user. Should be of the same length of generated-responses.
Text
text
string
The last input from the user in the conversation.
Parameters
Parameters.
Field
Field ID
Type
Note
Max Length
max-length
integer
Integer to define the maximum length in tokens of the output summary.
Max Time
max-time
number
The amount of time in seconds that the query should take maximum. Network can cause some overhead so it will be a soft limit.
Min Length
min-length
integer
Integer to define the minimum length in tokens of the output summary.
Repetition Penalty
repetition-penalty
number
The more a token is used within generation the more it is penalized to not be picked in successive generation passes.
Temperature
temperature
number
The temperature of the sampling operation. 1 means regular sampling, 0 means always take the highest score, 100.0 is getting closer to uniform probability.
Top K
top-k
integer
Integer to define the top tokens considered within the sample operation to create new text.
Top P
top-p
number
Float to define the tokens that are within the sample operation of text generation. Add tokens in the sample for more probable to least probable until the sum of the probabilities is greater than top-p.
Options
Options for the model.
Field
Field ID
Type
Note
Use Cache
use-cache
boolean
There is a cache layer on the inference API to speedup requests we have already seen. Most models can use those results as is as models are deterministic (meaning the results will be the same anyway). However if you use a non deterministic model, you can set this parameter to prevent the caching mechanism from being used resulting in a real new query.
Wait For Model
wait-for-model
boolean
If the model is not ready, wait for it instead of receiving 503. It limits the number of requests required to get your inference done. It is advised to only set this flag to true after receiving a 503 error as it will limit hanging in your application to known places.
A facility dictionary to send back for the next input (with the new user input addition).
Generated Text
generated-text
string
The answer of the bot.
Output Objects in Conversational
Conversation
Field
Field ID
Type
Note
Generated Responses
generated-responses
array
List of strings. The last outputs from the model in the conversation, after the model has run.
Past User Inputs
past-user-inputs
array
List of strings. The last inputs from the user in the conversation, after the model has run.
Image Classification
Image classification is the task of assigning a label or class to an entire image. Images are expected to have only one class for each image. Image classification models take an image as input and return a prediction about which class the image belongs to.
A float that represents how likely it is that the image file belongs to this class.
Image Segmentation
Image Segmentation divides an image into segments where each pixel in the image is mapped to an object. This task has multiple variants such as instance segmentation, panoptic segmentation and semantic segmentation.
The label for the class (model specific) of a segment.
Mask
mask
image/png
A str (base64 str of a single channel black-and-white img) representing the mask of a segment.
Score
score
number
A float that represents how likely it is that the segment belongs to the given class.
Object Detection
Object Detection models allow users to identify objects of certain defined classes. Object detection models receive an image as input and output the images with bounding boxes and labels on detected objects.
A dict (with keys [xmin,ymin,xmax,ymax]) representing the bounding box of a detected object.
Label
label
string
The label for the class (model specific) of a detected object.
Score
score
number
A float that represents how likely it is that the detected object belongs to the given class.
Box
Field
Field ID
Type
Note
X Max
xmax
number
X max.
X Min
xmin
number
X min.
Y Max
ymax
number
Y Max.
Y min
ymin
number
Y min.
Image to Text
Image to text models output a text from a given image. Image captioning or optical character recognition can be considered as the most common applications of image to text.
Input
Field ID
Type
Description
Task ID (required)
task
string
TASK_IMAGE_TO_TEXT
Model (required)
model
string
The Hugging Face model to be used.
Image (required)
image
string
The image file.
Output
Field ID
Type
Description
Text
text
string
Generated text.
Speech Recognition
Automatic Speech Recognition (ASR), also known as Speech to Text (STT), is the task of transcribing a given audio to text. It has many applications, such as voice user interfaces.
Input
Field ID
Type
Description
Task ID (required)
task
string
TASK_SPEECH_RECOGNITION
Model (required)
model
string
The Hugging Face model to be used.
Audio (required)
audio
string
The audio file.
Output
Field ID
Type
Description
Text
text
string
The string that was recognized within the audio file.
Audio Classification
Audio classification is the task of assigning a label or class to a given audio. It can be used for recognizing which command a user is giving or the emotion of a statement, as well as identifying a speaker.
