added remaining pipelines

.circleci/config.yml

@@ -18,7 +18,7 @@ jobs:
             . venv/bin/activate
             pip install --upgrade pip
             pip install -r gradio.egg-info/requires.txt
-            pip install shap IPython comet_ml wandb mlflow transformers
+            pip install shap IPython comet_ml wandb mlflow tensorflow transformers
             pip install selenium==4.0.0a6.post2 coverage scikit-image
       - run:
           command: |

gradio/external.py

@@ -90,21 +90,19 @@ def get_huggingface_interface(model_name, api_key, alias):
             'inputs': inputs.Textbox(label="Input"),
             'outputs': outputs.Label(label="Classification", type="confidences"),
             'preprocess': lambda x: {"inputs": x},
-            'postprocess': lambda r: {'Negative': r.json()[0][0]["score"],
-                                      'Positive': r.json()[0][1]["score"]}
-        },
-        'text2text-generation': {
-            'inputs': inputs.Textbox(label="Input"),
-            'outputs': outputs.Textbox(label="Generated Text"),
-            'preprocess': lambda x: {"inputs": x},
-            'postprocess': lambda r: r.json()[0]["generated_text"]
+            'postprocess': lambda r: {i["label"].split(", ")[0]: i["score"] for i in r.json()[0]}
         },
         'text-generation': {
             'inputs': inputs.Textbox(label="Input"),
             'outputs': outputs.Textbox(label="Output"),
             'preprocess': lambda x: {"inputs": x},
             'postprocess': lambda r: r.json()[0]["generated_text"],
-            # 'examples': [['My name is Clara and I am']]
+        },
+        'text2text-generation': {
+            'inputs': inputs.Textbox(label="Input"),
+            'outputs': outputs.Textbox(label="Generated Text"),
+            'preprocess': lambda x: {"inputs": x},
+            'postprocess': lambda r: r.json()[0]["generated_text"]
         },
         'translation': {
             'inputs': inputs.Textbox(label="Input"),
@@ -123,6 +121,7 @@ def get_huggingface_interface(model_name, api_key, alias):
             'postprocess': lambda r: {r.json()["labels"][i]: r.json()["scores"][i] for i in
                                       range(len(r.json()["labels"]))}
         },
+        # Non-HF pipelines
         'sentence-similarity': {
             # example model: hf.co/sentence-transformers/distilbert-base-nli-stsb-mean-tokens
             'inputs': [
@@ -166,9 +165,7 @@ def get_huggingface_interface(model_name, api_key, alias):
         response = requests.request("POST", api_url, headers=headers, data=data)        
         if not(response.status_code == 200):
             raise ValueError("Could not complete request to HuggingFace API, Error {}".format(response.status_code))
-        print("response>>>>>>", response.json())
         output = pipeline['postprocess'](response)
-        print("output>>>>>>", output)
         return output
     
     if alias is None:
@@ -295,9 +292,6 @@ def load_from_pipeline(pipeline):
             'preprocess': lambda i: {"images": i},
             'postprocess': lambda r: {i["label"].split(", ")[0]: i["score"] for i in r}
         }
-    elif hasattr(transformers, 'AutomaticSpeechRecognitionPipeline') and isinstance(pipeline, transformers.AutomaticSpeechRecognitionPipeline):
-        pipeline_info = {
-        }
     elif hasattr(transformers, 'QuestionAnsweringPipeline') and isinstance(pipeline, transformers.QuestionAnsweringPipeline):
         pipeline_info = {
             'inputs': [inputs.Textbox(label="Context", lines=7), inputs.Textbox(label="Question")],
@@ -305,13 +299,66 @@ def load_from_pipeline(pipeline):
             'preprocess': lambda c, q: {"context": c, "question": q},
             'postprocess': lambda r: (r["answer"], r["score"]),
         }
+    elif hasattr(transformers, 'SummarizationPipeline') and isinstance(pipeline, transformers.SummarizationPipeline):
+        pipeline_info = {
+            'inputs': inputs.Textbox(label="Input", lines=7),
+            'outputs': outputs.Textbox(label="Summary"),
+            'preprocess': lambda x: {"inputs": x},
+            'postprocess': lambda r: r[0]["summary_text"]
+        }
+    elif hasattr(transformers, 'TextClassificationPipeline') and isinstance(pipeline, transformers.TextClassificationPipeline):
+        pipeline_info = {
+            'inputs': inputs.Textbox(label="Input"),
+            'outputs': outputs.Label(label="Classification", type="confidences"),
+            'preprocess': lambda x: [x],
+            'postprocess': lambda r: {i["label"].split(", ")[0]: i["score"] for i in r}
+        }
+    elif hasattr(transformers, 'TextGenerationPipeline') and isinstance(pipeline, transformers.TextGenerationPipeline):
+        pipeline_info = {
+            'inputs': inputs.Textbox(label="Input"),
+            'outputs': outputs.Textbox(label="Output"),
+            'preprocess': lambda x: {"text_inputs": x},
+            'postprocess': lambda r: r[0]["generated_text"],
+        }
+    elif hasattr(transformers, 'TranslationPipeline') and isinstance(pipeline, transformers.TranslationPipeline):
+        pipeline_info = {
+            'inputs': inputs.Textbox(label="Input"),
+            'outputs': outputs.Textbox(label="Translation"),
+            'preprocess': lambda x: [x],
+            'postprocess': lambda r: r[0]["translation_text"]
+        }
+    elif hasattr(transformers, 'Text2TextGenerationPipeline') and isinstance(pipeline, transformers.Text2TextGenerationPipeline):
+        pipeline_info = {
+            'inputs': inputs.Textbox(label="Input"),
+            'outputs': outputs.Textbox(label="Generated Text"),
+            'preprocess': lambda x: [x],
+            'postprocess': lambda r: r[0]["generated_text"]
+        }
+    elif hasattr(transformers, 'ZeroShotClassificationPipeline') and isinstance(pipeline, transformers.ZeroShotClassificationPipeline):
+        pipeline_info = {
+            'inputs': [inputs.Textbox(label="Input"),
+                       inputs.Textbox(label="Possible class names ("
+                                            "comma-separated)"),
+                       inputs.Checkbox(label="Allow multiple true classes")],
+            'outputs': outputs.Label(label="Classification", type="confidences"),
+            'preprocess': lambda i, c, m: {"sequences": i, 
+                "candidate_labels": c, "multi_label": m},
+            'postprocess': lambda r: {r["labels"][i]: r["scores"][i] for i in
+                                      range(len(r["labels"]))}
+        }
     else:
         raise ValueError("Unsupported pipeline type: {}".format(type(pipeline)))
     
     # define the function that will be called by the Interface
     def fn(*params):
         data = pipeline_info["preprocess"](*params)
-        data = pipeline(**data)
+        # special cases that needs to be handled differently
+        if isinstance(pipeline, (transformers.TextClassificationPipeline, 
+                                 transformers.Text2TextGenerationPipeline,
+                                 transformers.TranslationPipeline)):
+            data = pipeline(*data)
+        else:
+            data = pipeline(**data)
         # print("Before postprocessing", data)
         output = pipeline_info["postprocess"](data)
         return output

test/test_external.py

@@ -155,7 +155,7 @@ class TestLoadInterface(unittest.TestCase):
         io = gr.Interface(**interface_info)
         io.api_mode = True
         output = io("I am happy, I love you.")
-        self.assertGreater(output['Positive'], 0.5)
+        self.assertGreater(output['POSITIVE'], 0.5)
 
     def test_image_classification_model(self):
         interface_info = gr.external.load_interface("models/google/vit-base-patch16-224")

test/test_interfaces.py

@@ -101,7 +101,7 @@ class TestInterface(unittest.TestCase):
     def test_interface_load(self):
         io = Interface.load("models/distilbert-base-uncased-finetuned-sst-2-english", alias="sentiment_classifier")
         output = io("I am happy, I love you.")
-        self.assertGreater(output['Positive'], 0.5)
+        self.assertGreater(output['POSITIVE'], 0.5)
         
     def test_interface_none_interp(self):
         interface = Interface(lambda x: x, "textbox", "label", interpretation=[None])