PyPI - edsl - Versions diffs - 0.1.36.dev6__py3-none-any.whl → 0.1.37__py3-none-any.whl - Mend

edsl 0.1.36.dev6py3-none-any.whl → 0.1.37py3-none-any.whl

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.

Files changed (261) hide show

edsl/Base.py +303 -303
edsl/BaseDiff.py +260 -260
edsl/TemplateLoader.py +24 -24
edsl/__init__.py +48 -47
edsl/__version__.py +1 -1
edsl/agents/Agent.py +855 -804
edsl/agents/AgentList.py +350 -337
edsl/agents/Invigilator.py +222 -222
edsl/agents/InvigilatorBase.py +284 -294
edsl/agents/PromptConstructor.py +353 -312
edsl/agents/__init__.py +3 -3
edsl/agents/descriptors.py +99 -86
edsl/agents/prompt_helpers.py +129 -129
edsl/auto/AutoStudy.py +117 -117
edsl/auto/StageBase.py +230 -230
edsl/auto/StageGenerateSurvey.py +178 -178
edsl/auto/StageLabelQuestions.py +125 -125
edsl/auto/StagePersona.py +61 -61
edsl/auto/StagePersonaDimensionValueRanges.py +88 -88
edsl/auto/StagePersonaDimensionValues.py +74 -74
edsl/auto/StagePersonaDimensions.py +69 -69
edsl/auto/StageQuestions.py +73 -73
edsl/auto/SurveyCreatorPipeline.py +21 -21
edsl/auto/utilities.py +224 -224
edsl/base/Base.py +289 -289
edsl/config.py +149 -149
edsl/conjure/AgentConstructionMixin.py +160 -152
edsl/conjure/Conjure.py +62 -62
edsl/conjure/InputData.py +659 -659
edsl/conjure/InputDataCSV.py +48 -48
edsl/conjure/InputDataMixinQuestionStats.py +182 -182
edsl/conjure/InputDataPyRead.py +91 -91
edsl/conjure/InputDataSPSS.py +8 -8
edsl/conjure/InputDataStata.py +8 -8
edsl/conjure/QuestionOptionMixin.py +76 -76
edsl/conjure/QuestionTypeMixin.py +23 -23
edsl/conjure/RawQuestion.py +65 -65
edsl/conjure/SurveyResponses.py +7 -7
edsl/conjure/__init__.py +9 -9
edsl/conjure/naming_utilities.py +263 -263
edsl/conjure/utilities.py +201 -201
edsl/conversation/Conversation.py +290 -238
edsl/conversation/car_buying.py +58 -58
edsl/conversation/chips.py +95 -0
edsl/conversation/mug_negotiation.py +81 -81
edsl/conversation/next_speaker_utilities.py +93 -93
edsl/coop/PriceFetcher.py +54 -54
edsl/coop/__init__.py +2 -2
edsl/coop/coop.py +958 -849
edsl/coop/utils.py +131 -131
edsl/data/Cache.py +527 -527
edsl/data/CacheEntry.py +228 -228
edsl/data/CacheHandler.py +149 -149
edsl/data/RemoteCacheSync.py +97 -84
edsl/data/SQLiteDict.py +292 -292
edsl/data/__init__.py +4 -4
edsl/data/orm.py +10 -10
edsl/data_transfer_models.py +73 -73
edsl/enums.py +173 -173
edsl/exceptions/BaseException.py +21 -0
edsl/exceptions/__init__.py +54 -50
edsl/exceptions/agents.py +38 -40
edsl/exceptions/configuration.py +16 -16
edsl/exceptions/coop.py +10 -10
edsl/exceptions/data.py +14 -14
edsl/exceptions/general.py +34 -34
edsl/exceptions/jobs.py +33 -33
edsl/exceptions/language_models.py +63 -63
edsl/exceptions/prompts.py +15 -15
edsl/exceptions/questions.py +91 -91
edsl/exceptions/results.py +29 -26
edsl/exceptions/scenarios.py +22 -0
edsl/exceptions/surveys.py +37 -34
edsl/inference_services/AnthropicService.py +87 -87
edsl/inference_services/AwsBedrock.py +120 -115
edsl/inference_services/AzureAI.py +217 -217
edsl/inference_services/DeepInfraService.py +18 -18
edsl/inference_services/GoogleService.py +156 -156
edsl/inference_services/GroqService.py +20 -20
edsl/inference_services/InferenceServiceABC.py +147 -147
edsl/inference_services/InferenceServicesCollection.py +97 -72
edsl/inference_services/MistralAIService.py +123 -123
edsl/inference_services/OllamaService.py +18 -18
edsl/inference_services/OpenAIService.py +224 -224
edsl/inference_services/TestService.py +89 -89
edsl/inference_services/TogetherAIService.py +170 -170
edsl/inference_services/models_available_cache.py +118 -118
edsl/inference_services/rate_limits_cache.py +25 -25
edsl/inference_services/registry.py +39 -39
edsl/inference_services/write_available.py +10 -10
edsl/jobs/Answers.py +56 -56
edsl/jobs/Jobs.py +1347 -1112
edsl/jobs/__init__.py +1 -1
edsl/jobs/buckets/BucketCollection.py +63 -63
edsl/jobs/buckets/ModelBuckets.py +65 -65
edsl/jobs/buckets/TokenBucket.py +248 -248
edsl/jobs/interviews/Interview.py +661 -651
edsl/jobs/interviews/InterviewExceptionCollection.py +99 -99
edsl/jobs/interviews/InterviewExceptionEntry.py +186 -182
edsl/jobs/interviews/InterviewStatistic.py +63 -63
edsl/jobs/interviews/InterviewStatisticsCollection.py +25 -25
edsl/jobs/interviews/InterviewStatusDictionary.py +78 -78
edsl/jobs/interviews/InterviewStatusLog.py +92 -92
edsl/jobs/interviews/ReportErrors.py +66 -66
edsl/jobs/interviews/interview_status_enum.py +9 -9
edsl/jobs/runners/JobsRunnerAsyncio.py +338 -337
edsl/jobs/runners/JobsRunnerStatus.py +332 -332
edsl/jobs/tasks/QuestionTaskCreator.py +242 -242
edsl/jobs/tasks/TaskCreators.py +64 -64
edsl/jobs/tasks/TaskHistory.py +442 -441
edsl/jobs/tasks/TaskStatusLog.py +23 -23
edsl/jobs/tasks/task_status_enum.py +163 -163
edsl/jobs/tokens/InterviewTokenUsage.py +27 -27
edsl/jobs/tokens/TokenUsage.py +34 -34
edsl/language_models/KeyLookup.py +30 -0
edsl/language_models/LanguageModel.py +706 -718
edsl/language_models/ModelList.py +102 -102
edsl/language_models/RegisterLanguageModelsMeta.py +184 -184
edsl/language_models/__init__.py +3 -2
edsl/language_models/fake_openai_call.py +15 -15
edsl/language_models/fake_openai_service.py +61 -61
edsl/language_models/registry.py +137 -137
edsl/language_models/repair.py +156 -156
edsl/language_models/unused/ReplicateBase.py +83 -83
edsl/language_models/utilities.py +64 -64
edsl/notebooks/Notebook.py +259 -259
edsl/notebooks/__init__.py +1 -1
edsl/prompts/Prompt.py +357 -358
edsl/prompts/__init__.py +2 -2
edsl/questions/AnswerValidatorMixin.py +289 -289
edsl/questions/QuestionBase.py +656 -616
edsl/questions/QuestionBaseGenMixin.py +161 -161
edsl/questions/QuestionBasePromptsMixin.py +234 -266
edsl/questions/QuestionBudget.py +227 -227
edsl/questions/QuestionCheckBox.py +359 -359
edsl/questions/QuestionExtract.py +183 -183
edsl/questions/QuestionFreeText.py +114 -113
edsl/questions/QuestionFunctional.py +159 -159
edsl/questions/QuestionList.py +231 -231
edsl/questions/QuestionMultipleChoice.py +286 -286
edsl/questions/QuestionNumerical.py +153 -153
edsl/questions/QuestionRank.py +324 -324
edsl/questions/Quick.py +41 -41
edsl/questions/RegisterQuestionsMeta.py +71 -71
edsl/questions/ResponseValidatorABC.py +174 -174
edsl/questions/SimpleAskMixin.py +73 -73
edsl/questions/__init__.py +26 -26
edsl/questions/compose_questions.py +98 -98
edsl/questions/decorators.py +21 -21
edsl/questions/derived/QuestionLikertFive.py +76 -76
edsl/questions/derived/QuestionLinearScale.py +87 -87
edsl/questions/derived/QuestionTopK.py +91 -91
edsl/questions/derived/QuestionYesNo.py +82 -82
edsl/questions/descriptors.py +413 -418
edsl/questions/prompt_templates/question_budget.jinja +13 -13
edsl/questions/prompt_templates/question_checkbox.jinja +32 -32
edsl/questions/prompt_templates/question_extract.jinja +11 -11
edsl/questions/prompt_templates/question_free_text.jinja +3 -3
edsl/questions/prompt_templates/question_linear_scale.jinja +11 -11
edsl/questions/prompt_templates/question_list.jinja +17 -17
edsl/questions/prompt_templates/question_multiple_choice.jinja +33 -33
edsl/questions/prompt_templates/question_numerical.jinja +36 -36
edsl/questions/question_registry.py +147 -147
edsl/questions/settings.py +12 -12
edsl/questions/templates/budget/answering_instructions.jinja +7 -7
edsl/questions/templates/budget/question_presentation.jinja +7 -7
edsl/questions/templates/checkbox/answering_instructions.jinja +10 -10
edsl/questions/templates/checkbox/question_presentation.jinja +22 -22
edsl/questions/templates/extract/answering_instructions.jinja +7 -7
edsl/questions/templates/likert_five/answering_instructions.jinja +10 -10
edsl/questions/templates/likert_five/question_presentation.jinja +11 -11
edsl/questions/templates/linear_scale/answering_instructions.jinja +5 -5
edsl/questions/templates/linear_scale/question_presentation.jinja +5 -5
edsl/questions/templates/list/answering_instructions.jinja +3 -3
edsl/questions/templates/list/question_presentation.jinja +5 -5
edsl/questions/templates/multiple_choice/answering_instructions.jinja +9 -9
edsl/questions/templates/multiple_choice/question_presentation.jinja +11 -11
edsl/questions/templates/numerical/answering_instructions.jinja +6 -6
edsl/questions/templates/numerical/question_presentation.jinja +6 -6
edsl/questions/templates/rank/answering_instructions.jinja +11 -11
edsl/questions/templates/rank/question_presentation.jinja +15 -15
edsl/questions/templates/top_k/answering_instructions.jinja +8 -8
edsl/questions/templates/top_k/question_presentation.jinja +22 -22
edsl/questions/templates/yes_no/answering_instructions.jinja +6 -6
edsl/questions/templates/yes_no/question_presentation.jinja +11 -11
edsl/results/Dataset.py +293 -293
edsl/results/DatasetExportMixin.py +717 -693
edsl/results/DatasetTree.py +145 -145
edsl/results/Result.py +450 -433
edsl/results/Results.py +1071 -1158
edsl/results/ResultsDBMixin.py +238 -238
edsl/results/ResultsExportMixin.py +43 -43
edsl/results/ResultsFetchMixin.py +33 -33
edsl/results/ResultsGGMixin.py +121 -121
edsl/results/ResultsToolsMixin.py +98 -98
edsl/results/Selector.py +135 -118
edsl/results/__init__.py +2 -2
edsl/results/tree_explore.py +115 -115
edsl/scenarios/FileStore.py +458 -443
edsl/scenarios/Scenario.py +546 -507
edsl/scenarios/ScenarioHtmlMixin.py +64 -59
edsl/scenarios/ScenarioList.py +1112 -1101
edsl/scenarios/ScenarioListExportMixin.py +52 -52
edsl/scenarios/ScenarioListPdfMixin.py +261 -261
edsl/scenarios/__init__.py +4 -2
edsl/shared.py +1 -1
edsl/study/ObjectEntry.py +173 -173
edsl/study/ProofOfWork.py +113 -113
edsl/study/SnapShot.py +80 -80
edsl/study/Study.py +528 -528
edsl/study/__init__.py +4 -4
edsl/surveys/DAG.py +148 -148
edsl/surveys/Memory.py +31 -31
edsl/surveys/MemoryPlan.py +244 -244
edsl/surveys/Rule.py +330 -324
edsl/surveys/RuleCollection.py +387 -387
edsl/surveys/Survey.py +1795 -1772
edsl/surveys/SurveyCSS.py +261 -261
edsl/surveys/SurveyExportMixin.py +259 -259
edsl/surveys/SurveyFlowVisualizationMixin.py +121 -121
edsl/surveys/SurveyQualtricsImport.py +284 -284
edsl/surveys/__init__.py +3 -3
edsl/surveys/base.py +53 -53
edsl/surveys/descriptors.py +56 -56
edsl/surveys/instructions/ChangeInstruction.py +47 -47
edsl/surveys/instructions/Instruction.py +51 -51
edsl/surveys/instructions/InstructionCollection.py +77 -77
edsl/templates/error_reporting/base.html +23 -23
edsl/templates/error_reporting/exceptions_by_model.html +34 -34
edsl/templates/error_reporting/exceptions_by_question_name.html +16 -16
edsl/templates/error_reporting/exceptions_by_type.html +16 -16
edsl/templates/error_reporting/interview_details.html +115 -115
edsl/templates/error_reporting/interviews.html +9 -9
edsl/templates/error_reporting/overview.html +4 -4
edsl/templates/error_reporting/performance_plot.html +1 -1
edsl/templates/error_reporting/report.css +73 -73
edsl/templates/error_reporting/report.html +117 -117
edsl/templates/error_reporting/report.js +25 -25
edsl/tools/__init__.py +1 -1
edsl/tools/clusters.py +192 -192
edsl/tools/embeddings.py +27 -27
edsl/tools/embeddings_plotting.py +118 -118
edsl/tools/plotting.py +112 -112
edsl/tools/summarize.py +18 -18
edsl/utilities/SystemInfo.py +28 -28
edsl/utilities/__init__.py +22 -22
edsl/utilities/ast_utilities.py +25 -25
edsl/utilities/data/Registry.py +6 -6
edsl/utilities/data/__init__.py +1 -1
edsl/utilities/data/scooter_results.json +1 -1
edsl/utilities/decorators.py +77 -77
edsl/utilities/gcp_bucket/cloud_storage.py +96 -96
edsl/utilities/interface.py +627 -627
edsl/utilities/repair_functions.py +28 -28
edsl/utilities/restricted_python.py +70 -70
edsl/utilities/utilities.py +409 -391
{edsl-0.1.36.dev6.dist-info → edsl-0.1.37.dist-info}/LICENSE +21 -21
{edsl-0.1.36.dev6.dist-info → edsl-0.1.37.dist-info}/METADATA +1 -1
edsl-0.1.37.dist-info/RECORD +283 -0
edsl-0.1.36.dev6.dist-info/RECORD +0 -279
{edsl-0.1.36.dev6.dist-info → edsl-0.1.37.dist-info}/WHEEL +0 -0

edsl/tools/clusters.py CHANGED Viewed

@@ -1,192 +1,192 @@
-import json
-import numpy as np
-from sklearn.cluster import KMeans
-from sklearn.manifold import TSNE
-from sklearn.decomposition import PCA
-from IPython.display import display_html
-def compute_tsne(embeddings, cluster_labels, text_labels):
-    """
-    Compute t-SNE on embedding vectors.
-    Parameters:
-    embeddings (np.ndarray): The embedding vectors.
-    cluster_labels (np.ndarray): Cluster labels for each embedding.
-    text_labels (list): Text labels for each embedding.
-    Returns:
-    list: List of dictionaries with x, y coordinates, cluster labels, and text labels.
-    """
-    tsne = TSNE(n_components=2, random_state=42)
-    tsne_results = tsne.fit_transform(embeddings)
-    data = [
-        {
-            "x": float(tsne_results[i, 0]),
-            "y": float(tsne_results[i, 1]),
-            "cluster_label": str(cluster_labels[i]),
-            "text_label": text_labels[i],
-        }
-        for i in range(len(cluster_labels))
-    ]
-    return data
-def compute_pca(embeddings, cluster_labels, text_labels):
-    """
-    Compute PCA on embedding vectors.
-    Parameters:
-    embeddings (np.ndarray): The embedding vectors.
-    cluster_labels (np.ndarray): Cluster labels for each embedding.
-    text_labels (list): Text labels for each embedding.
-    Returns:
-    list: List of dictionaries with x, y coordinates, cluster labels, and text labels.
-    """
-    pca = PCA(n_components=2)
-    pca_results = pca.fit_transform(embeddings)
-    data = [
-        {
-            "x": float(pca_results[i, 0]),
-            "y": float(pca_results[i, 1]),
-            "cluster_label": str(cluster_labels[i]),
-            "text_label": text_labels[i],
-        }
-        for i in range(len(cluster_labels))
-    ]
-    return data
-def plot(embeddings, text_labels, n_clusters=5, method="tsne"):
-    """
-    Perform k-means clustering and plot results in a Jupyter notebook using D3.js.
-    Parameters:
-    embeddings (np.ndarray): The embedding vectors.
-    text_labels (list): Text labels for each embedding.
-    n_clusters (int): The number of clusters to form.
-    method (str): The dimensionality reduction method to use ('tsne' or 'pca').
-    """
-    # Perform k-means clustering
-    kmeans = KMeans(n_clusters=n_clusters, random_state=42)
-    cluster_labels = kmeans.fit_predict(embeddings)
-    # Compute dimensionality reduction
-    if method == "tsne":
-        data = compute_tsne(embeddings, cluster_labels, text_labels)
-    elif method == "pca":
-        data = compute_pca(embeddings, cluster_labels, text_labels)
-    else:
-        raise ValueError("Invalid method. Choose 'tsne' or 'pca'.")
-    # Convert data to JSON
-    data_json = json.dumps(data)
-    # HTML content with embedded data
-    html_content = f"""
-    <!DOCTYPE html>
-    <html lang="en">
-    <head>
-        <meta charset="UTF-8">
-        <title>t-SNE/PCA Plot with D3.js</title>
-        <script src="https://d3js.org/d3.v6.min.js"></script>
-        <style>
-            .tooltip {{
-                position: absolute;
-                text-align: center;
-                width: auto;
-                height: auto;
-                padding: 2px;
-                font: 12px sans-serif;
-                background: lightsteelblue;
-                border: 0px;
-                border-radius: 8px;
-                pointer-events: none;
-            }}
-            .dot {{
-                stroke: #000;
-                stroke-width: 0.5;
-            }}
-        </style>
-    </head>
-    <body>
-        <svg width="600" height="600"></svg>
-        <script>
-            // Embedded data
-            const data = {data_json};
-            const svg = d3.select("svg"),
-                width = +svg.attr("width"),
-                height = +svg.attr("height");
-            // Set up scales
-            const x = d3.scaleLinear()
-                .domain(d3.extent(data, d => d.x))
-                .range([0, width]);
-            const y = d3.scaleLinear()
-                .domain(d3.extent(data, d => d.y))
-                .range([height, 0]);
-            // Set up color scale
-            const color = d3.scaleOrdinal(d3.schemeCategory10);
-            // Create tooltip
-            const tooltip = d3.select("body").append("div")
-                .attr("class", "tooltip")
-                .style("opacity", 0);
-            // Create circles for each point
-            svg.selectAll("circle")
-                .data(data)
-                .enter().append("circle")
-                .attr("cx", d => x(d.x))
-                .attr("cy", d => y(d.y))
-                .attr("r", 5)
-                .attr("class", "dot")
-                .style("fill", d => color(d.cluster_label))
-                .on("mouseover", function(event, d) {{
-                    tooltip.transition()
-                        .duration(200)
-                        .style("opacity", .9);
-                    tooltip.html(d.text_label)
-                        .style("left", (event.pageX + 5) + "px")
-                        .style("top", (event.pageY - 28) + "px");
-                }})
-                .on("mouseout", function(d) {{
-                    tooltip.transition()
-                        .duration(500)
-                        .style("opacity", 0);
-                }});
-        </script>
-    </body>
-    </html>
-    """
-    # Write HTML content to a temporary file
-    html_file = "tsne_pca_plot.html"
-    with open(html_file, "w") as file:
-        file.write(html_content)
-    # Display the HTML content in an iframe within a Jupyter notebook
-    display_html(
-        f'<iframe src="{html_file}" width="600" height="600"></iframe>', raw=True
-    )
-# Example usage
-if __name__ == "__main__":
-    # Generate some sample data (embedding vectors)
-    np.random.seed(42)
-    embedding_vectors = np.random.rand(
-        100, 50
-    )  # 100 samples with 50-dimensional embeddings
-    text_labels = [f"Text {i}" for i in range(100)]  # Sample text labels
-    # Plot the clusters using t-SNE
-    plot(embedding_vectors, text_labels, n_clusters=5, method="tsne")
-    # Plot the clusters using PCA
-    plot(embedding_vectors, text_labels, n_clusters=5, method="pca")
+import json
+import numpy as np
+from sklearn.cluster import KMeans
+from sklearn.manifold import TSNE
+from sklearn.decomposition import PCA
+from IPython.display import display_html
+def compute_tsne(embeddings, cluster_labels, text_labels):
+    """
+    Compute t-SNE on embedding vectors.
+    Parameters:
+    embeddings (np.ndarray): The embedding vectors.
+    cluster_labels (np.ndarray): Cluster labels for each embedding.
+    text_labels (list): Text labels for each embedding.
+    Returns:
+    list: List of dictionaries with x, y coordinates, cluster labels, and text labels.
+    """
+    tsne = TSNE(n_components=2, random_state=42)
+    tsne_results = tsne.fit_transform(embeddings)
+    data = [
+        {
+            "x": float(tsne_results[i, 0]),
+            "y": float(tsne_results[i, 1]),
+            "cluster_label": str(cluster_labels[i]),
+            "text_label": text_labels[i],
+        }
+        for i in range(len(cluster_labels))
+    ]
+    return data
+def compute_pca(embeddings, cluster_labels, text_labels):
+    """
+    Compute PCA on embedding vectors.
+    Parameters:
+    embeddings (np.ndarray): The embedding vectors.
+    cluster_labels (np.ndarray): Cluster labels for each embedding.
+    text_labels (list): Text labels for each embedding.
+    Returns:
+    list: List of dictionaries with x, y coordinates, cluster labels, and text labels.
+    """
+    pca = PCA(n_components=2)
+    pca_results = pca.fit_transform(embeddings)
+    data = [
+        {
+            "x": float(pca_results[i, 0]),
+            "y": float(pca_results[i, 1]),
+            "cluster_label": str(cluster_labels[i]),
+            "text_label": text_labels[i],
+        }
+        for i in range(len(cluster_labels))
+    ]
+    return data
+def plot(embeddings, text_labels, n_clusters=5, method="tsne"):
+    """
+    Perform k-means clustering and plot results in a Jupyter notebook using D3.js.
+    Parameters:
+    embeddings (np.ndarray): The embedding vectors.
+    text_labels (list): Text labels for each embedding.
+    n_clusters (int): The number of clusters to form.
+    method (str): The dimensionality reduction method to use ('tsne' or 'pca').
+    """
+    # Perform k-means clustering
+    kmeans = KMeans(n_clusters=n_clusters, random_state=42)
+    cluster_labels = kmeans.fit_predict(embeddings)
+    # Compute dimensionality reduction
+    if method == "tsne":
+        data = compute_tsne(embeddings, cluster_labels, text_labels)
+    elif method == "pca":
+        data = compute_pca(embeddings, cluster_labels, text_labels)
+    else:
+        raise ValueError("Invalid method. Choose 'tsne' or 'pca'.")
+    # Convert data to JSON
+    data_json = json.dumps(data)
+    # HTML content with embedded data
+    html_content = f"""
+    <!DOCTYPE html>
+    <html lang="en">
+    <head>
+        <meta charset="UTF-8">
+        <title>t-SNE/PCA Plot with D3.js</title>
+        <script src="https://d3js.org/d3.v6.min.js"></script>
+        <style>
+            .tooltip {{
+                position: absolute;
+                text-align: center;
+                width: auto;
+                height: auto;
+                padding: 2px;
+                font: 12px sans-serif;
+                background: lightsteelblue;
+                border: 0px;
+                border-radius: 8px;
+                pointer-events: none;
+            }}
+            .dot {{
+                stroke: #000;
+                stroke-width: 0.5;
+            }}
+        </style>
+    </head>
+    <body>
+        <svg width="600" height="600"></svg>
+        <script>
+            // Embedded data
+            const data = {data_json};
+            const svg = d3.select("svg"),
+                width = +svg.attr("width"),
+                height = +svg.attr("height");
+            // Set up scales
+            const x = d3.scaleLinear()
+                .domain(d3.extent(data, d => d.x))
+                .range([0, width]);
+            const y = d3.scaleLinear()
+                .domain(d3.extent(data, d => d.y))
+                .range([height, 0]);
+            // Set up color scale
+            const color = d3.scaleOrdinal(d3.schemeCategory10);
+            // Create tooltip
+            const tooltip = d3.select("body").append("div")
+                .attr("class", "tooltip")
+                .style("opacity", 0);
+            // Create circles for each point
+            svg.selectAll("circle")
+                .data(data)
+                .enter().append("circle")
+                .attr("cx", d => x(d.x))
+                .attr("cy", d => y(d.y))
+                .attr("r", 5)
+                .attr("class", "dot")
+                .style("fill", d => color(d.cluster_label))
+                .on("mouseover", function(event, d) {{
+                    tooltip.transition()
+                        .duration(200)
+                        .style("opacity", .9);
+                    tooltip.html(d.text_label)
+                        .style("left", (event.pageX + 5) + "px")
+                        .style("top", (event.pageY - 28) + "px");
+                }})
+                .on("mouseout", function(d) {{
+                    tooltip.transition()
+                        .duration(500)
+                        .style("opacity", 0);
+                }});
+        </script>
+    </body>
+    </html>
+    """
+    # Write HTML content to a temporary file
+    html_file = "tsne_pca_plot.html"
+    with open(html_file, "w") as file:
+        file.write(html_content)
+    # Display the HTML content in an iframe within a Jupyter notebook
+    display_html(
+        f'<iframe src="{html_file}" width="600" height="600"></iframe>', raw=True
+    )
+# Example usage
+if __name__ == "__main__":
+    # Generate some sample data (embedding vectors)
+    np.random.seed(42)
+    embedding_vectors = np.random.rand(
+        100, 50
+    )  # 100 samples with 50-dimensional embeddings
+    text_labels = [f"Text {i}" for i in range(100)]  # Sample text labels
+    # Plot the clusters using t-SNE
+    plot(embedding_vectors, text_labels, n_clusters=5, method="tsne")
+    # Plot the clusters using PCA
+    plot(embedding_vectors, text_labels, n_clusters=5, method="pca")

edsl/tools/embeddings.py CHANGED Viewed

@@ -1,27 +1,27 @@
-import openai
-from dotenv import load_dotenv
-import os
-# Load environment variables from .env file
-load_dotenv()
-# Get the OpenAI API key from the environment variable
-openai.api_key = os.getenv("OPENAI_API_KEY")
-from openai import Client
-def get_embeddings(texts):
-    """
-    Get embeddings for a list of texts using OpenAI API.
-    Args:
-        texts (list of str): List of strings to get embeddings for.
-    Returns:
-        list of list of float: List of embeddings.
-    """
-    client = Client()
-    response = client.embeddings.create(input=texts, model="text-embedding-ada-002")
-    embeddings = [item.embedding for item in response.data]
-    return embeddings
+import openai
+from dotenv import load_dotenv
+import os
+# Load environment variables from .env file
+load_dotenv()
+# Get the OpenAI API key from the environment variable
+openai.api_key = os.getenv("OPENAI_API_KEY")
+from openai import Client
+def get_embeddings(texts):
+    """
+    Get embeddings for a list of texts using OpenAI API.
+    Args:
+        texts (list of str): List of strings to get embeddings for.
+    Returns:
+        list of list of float: List of embeddings.
+    """
+    client = Client()
+    response = client.embeddings.create(input=texts, model="text-embedding-ada-002")
+    embeddings = [item.embedding for item in response.data]
+    return embeddings

edsl 0.1.36.dev6__py3-none-any.whl → 0.1.37__py3-none-any.whl

edsl 0.1.36.dev6py3-none-any.whl → 0.1.37py3-none-any.whl